Tag Archives: ECS

GPU workloads on AWS Batch

Post Syndicated from Josh Rad original https://aws.amazon.com/blogs/compute/gpu-workloads-on-aws-batch/

Contributed by Manuel Manzano Hoss, Cloud Support Engineer

I remember playing around with graphics processing units (GPUs) workload examples in 2017 when the Deep Learning on AWS Batch post was published by my colleague Kiuk Chung. He provided an example of how to train a convolutional neural network (CNN), the LeNet architecture, to recognize handwritten digits from the MNIST dataset using Apache MXNet as the framework. Back then, to run such jobs with GPU capabilities, I had to do the following:

  • Create a custom GPU-enabled AMI that had installed Docker, the ECS agent, NVIDIA driver and container runtime, and CUDA.
  • Identify the type of P2 EC2 instance that had the required amount of GPU for my job.
  • Check the amount of vCPUs that it offered (even if I was not interested on using them).
  • Specify that number of vCPUs for my job.

All that, when I didn’t have any certainty that the instance was going to have the GPU required available when my job was already running. Back then, there was no GPU pinning. Other jobs running on the same EC2 instance were able to use that GPU, making the orchestration of my jobs a tricky task.

Fast forward two years. Today, AWS Batch announced integrated support for Amazon EC2 Accelerated Instances. It is now possible to specify an amount of GPU as a resource that AWS Batch considers in choosing the EC2 instance to run your job, along with vCPU and memory. That allows me to take advantage of the main benefits of using AWS Batch, the compute resource selection algorithm and job scheduler. It also frees me from having to check the types of EC2 instances that have enough GPU.

Also, I can take advantage of the Amazon ECS GPU-optimized AMI maintained by AWS. It comes with the NVIDIA drivers and all the necessary software to run GPU-enabled jobs. When I allow the P2 or P3 instance types on my compute environment, AWS Batch launches my compute resources using the Amazon ECS GPU-optimized AMI automatically.

In other words, now I don’t worry about the GPU task list mentioned earlier. I can focus on deciding which framework and command to run on my GPU-accelerated workload. At the same time, I’m now sure that my jobs have access to the required performance, as physical GPUs are pinned to each job and not shared among them.

A GPU race against the past

As a kind of GPU-race exercise, I checked a similar example to the one from Kiuk’s post, to see how fast it could be to run a GPU-enabled job now. I used the AWS Management Console to demonstrate how simple the steps are.

In this case, I decided to use the deep neural network architecture called multilayer perceptron (MLP), not the LeNet CNN, to compare the validation accuracy between them.

To make the test even simpler and faster to implement, I thought I would use one of the recently announced AWS Deep Learning containers, which come pre-packed with different frameworks and ready-to-process data. I chose the container that comes with MXNet and Python 2.7, customized for Training and GPU. For more information about the Docker images available, see the AWS Deep Learning Containers documentation.

In the AWS Batch console, I created a managed compute environment with the default settings, allowing AWS Batch to create the required IAM roles on my behalf.

On the configuration of the compute resources, I selected the P2 and P3 families of instances, as those are the type of instance with GPU capabilities. You can select On-Demand Instances, but in this case I decided to use Spot Instances to take advantage of the discounts that this pricing model offers. I left the defaults for all other settings, selecting the AmazonEC2SpotFleetRole role that I created the first time that I used Spot Instances:

Finally, I also left the network settings as default. My compute environment selected the default VPC, three subnets, and a security group. They are enough to run my jobs and at the same time keep my environment safe by limiting connections from outside the VPC:

I created a job queue, GPU_JobQueue, attaching it to the compute environment that I just created:

Next, I registered the same job definition that I would have created following Kiuk’s post. I specified enough memory to run this test, one vCPU, and the AWS Deep Learning Docker image that I chose, in this case mxnet-training:1.4.0-gpu-py27-cu90-ubuntu16.04. The amount of GPU required was in this case, one. To have access to run the script, the container must run as privileged, or using the root user.

Finally, I submitted the job. I first cloned the MXNet repository for the train_mnist.py Python script. Then I ran the script itself, with the parameter –gpus 0 to indicate that the assigned GPU should be used. The job inherits all the other parameters from the job definition:

sh -c 'git clone -b 1.3.1 https://github.com/apache/incubator-mxnet.git && python /incubator-mxnet/example/image-classification/train_mnist.py --gpus 0'

That’s all, and my GPU-enabled job was running. It took me less than two minutes to go from zero to having the job submitted. This is the log of my job, from which I removed the iterations from epoch 1 to 18 to make it shorter:

14:32:31     Cloning into 'incubator-mxnet'...

14:33:50     Note: checking out '19c501680183237d52a862e6ae1dc4ddc296305b'.

14:33:51     INFO:root:start with arguments Namespace(add_stn=False, batch_size=64, disp_batches=100, dtype='float32', gc_threshold=0.5, gc_type='none', gpus='0', initializer='default', kv_store='device', load_epoch=None, loss='', lr=0.05, lr_factor=0.1, lr_step_epochs='10', macrobatch_size=0, model_prefix=None, mom=0.9, monitor=0, network='mlp', num_classes=10, num_epochs=20, num_examples=60000, num_layers=No

14:33:51     DEBUG:urllib3.connectionpool:Starting new HTTP connection (1): yann.lecun.com:80

14:33:54     DEBUG:urllib3.connectionpool:http://yann.lecun.com:80 "GET /exdb/mnist/train-labels-idx1-ubyte.gz HTTP/1.1" 200 28881

14:33:55     DEBUG:urllib3.connectionpool:Starting new HTTP connection (1): yann.lecun.com:80

14:33:55     DEBUG:urllib3.connectionpool:http://yann.lecun.com:80 "GET /exdb/mnist/train-images-idx3-ubyte.gz HTTP/1.1" 200 9912422

14:33:59     DEBUG:urllib3.connectionpool:Starting new HTTP connection (1): yann.lecun.com:80

14:33:59     DEBUG:urllib3.connectionpool:http://yann.lecun.com:80 "GET /exdb/mnist/t10k-labels-idx1-ubyte.gz HTTP/1.1" 200 4542

14:33:59     DEBUG:urllib3.connectionpool:Starting new HTTP connection (1): yann.lecun.com:80

14:34:00     DEBUG:urllib3.connectionpool:http://yann.lecun.com:80 "GET /exdb/mnist/t10k-images-idx3-ubyte.gz HTTP/1.1" 200 1648877

14:34:04     INFO:root:Epoch[0] Batch [0-100] Speed: 37038.30 samples/sec accuracy=0.793472

14:34:04     INFO:root:Epoch[0] Batch [100-200] Speed: 36457.89 samples/sec accuracy=0.906719

14:34:04     INFO:root:Epoch[0] Batch [200-300] Speed: 36981.20 samples/sec accuracy=0.927500

14:34:04     INFO:root:Epoch[0] Batch [300-400] Speed: 36925.04 samples/sec accuracy=0.935156

14:34:04     INFO:root:Epoch[0] Batch [400-500] Speed: 37262.36 samples/sec accuracy=0.940156

14:34:05     INFO:root:Epoch[0] Batch [500-600] Speed: 37729.64 samples/sec accuracy=0.942813

14:34:05     INFO:root:Epoch[0] Batch [600-700] Speed: 37493.55 samples/sec accuracy=0.949063

14:34:05     INFO:root:Epoch[0] Batch [700-800] Speed: 37320.80 samples/sec accuracy=0.953906

14:34:05     INFO:root:Epoch[0] Batch [800-900] Speed: 37705.85 samples/sec accuracy=0.958281

14:34:05     INFO:root:Epoch[0] Train-accuracy=0.924024

14:34:05     INFO:root:Epoch[0] Time cost=1.633

...  LOGS REMOVED

14:34:44     INFO:root:Epoch[19] Batch [0-100] Speed: 36864.44 samples/sec accuracy=0.999691

14:34:44     INFO:root:Epoch[19] Batch [100-200] Speed: 37088.35 samples/sec accuracy=1.000000

14:34:44     INFO:root:Epoch[19] Batch [200-300] Speed: 36706.91 samples/sec accuracy=0.999687

14:34:44     INFO:root:Epoch[19] Batch [300-400] Speed: 37941.19 samples/sec accuracy=0.999687

14:34:44     INFO:root:Epoch[19] Batch [400-500] Speed: 37180.97 samples/sec accuracy=0.999844

14:34:44     INFO:root:Epoch[19] Batch [500-600] Speed: 37122.30 samples/sec accuracy=0.999844

14:34:45     INFO:root:Epoch[19] Batch [600-700] Speed: 37199.37 samples/sec accuracy=0.999687

14:34:45     INFO:root:Epoch[19] Batch [700-800] Speed: 37284.93 samples/sec accuracy=0.999219

14:34:45     INFO:root:Epoch[19] Batch [800-900] Speed: 36996.80 samples/sec accuracy=0.999844

14:34:45     INFO:root:Epoch[19] Train-accuracy=0.999733

14:34:45     INFO:root:Epoch[19] Time cost=1.617

14:34:45     INFO:root:Epoch[19] Validation-accuracy=0.983579

Summary

As you can see, after AWS Batch launched the instance, the job took slightly more than two minutes to run. I spent roughly five minutes from start to finish. That was much faster than the time that I was previously spending just to configure the AMI. Using the AWS CLI, one of the AWS SDKs, or AWS CloudFormation, the same environment could be created even faster.

From a training point of view, I lost on the validation accuracy, as the results obtained using the LeNet CNN are higher than when using an MLP network. On the other hand, my job was faster, with a time cost of 1.6 seconds in average for each epoch. As the software stack evolves, and increased hardware capabilities come along, these numbers keep improving, but that shouldn’t mean extra complexity. Using managed primitives like the one presented in this post enables a simpler implementation.

I encourage you to test this example and see for yourself how just a few clicks or commands lets you start running GPU jobs with AWS Batch. Then, it is just a matter of replacing the Docker image that I used for one with the framework of your choice, TensorFlow, Caffe, PyTorch, Keras, etc. Start to run your GPU-enabled machine learning, deep learning, computational fluid dynamics (CFD), seismic analysis, molecular modeling, genomics, or computational finance workloads. It’s faster and easier than ever.

If you decide to give it a try, have any doubt or just want to let me know what you think about this post, please write in the comments section!

Amazon ECS Task Placement

Post Syndicated from tiffany jernigan (@tiffanyfayj) original https://aws.amazon.com/blogs/compute/amazon-ecs-task-placement/

Intro

Amazon Elastic Container Service (ECS) is a highly scalable, high-performance container orchestration service that allows you to easily run and scale containerized applications on AWS. This post covers how Amazon Elastic Container Service (Amazon ECS) runs containers in a cluster. Topics include why AWS built the task placement engine, the different strategies and constraints available to decide where and how containers are run, and things to consider when picking placement strategies.

If you are not familiar with the relationship between ECS and Amazon EC2 or its components, see the Building Blocks of Amazon ECS post.

Task Placement

When a task is launched in a cluster, a decision has to be made to choose which container instance should run that task. Conversely, when scaling down a service, a decision has to be made to choose the specific task to be terminated.

Task placement

By default, ECS uses the following placement strategies:

  • When you run tasks with the RunTask API action, tasks are placed randomly in a cluster.
  • When you launch and terminate tasks with the CreateService API action, the service scheduler spreads the tasks across the Availability Zones (and the instances within the zones) in a cluster.

Before December 2016, tasks could only be placed by their default placement strategies. This meant making the decision yourself, such as writing your own scheduler, and calling the StartTask API action to achieve custom task placement. When you manually constrained the placement of your grouping of containers, you could only place based on CPU, memory, and ports. Additionally, while creating your own scheduler can be powerful, there’s a tradeoff with complexity.

AWS built the task placement engine, which removes the need for you to build, run, and manage your own scheduling and placement services. There are several new features that provide you with more control over how applications run across clusters through custom attributes.

You can think of this flow as a funnel with filters for your instances. Constraints must be obeyed. If an instance doesn’t fit, it isn’t used. Strategies are then used to sort the rest of the instances by preference to determine which are the “best.”

For every instantiation of your task, it runs through every step. Calling run-task with a count of n is effectively calling run-task n times (create-service also works the same way).

Cluster Constraints, Placement Constraints, Placement Strategies

Example

Here’s how to use these placement features. In this example, you use the AWS CLI run-task command. For the last couple of filters, I show how to use them with placement flags, but you can just as easily include them in your task definition file instead. This can all be done in the console as well. Start with the cluster shown earlier:

Task Placement Instances

aws ecs run-task --task-definition nouvelleApp \
--placement-constraints type="memberOf",expression="attribute:ecs.instance-type == t2.small" \
--placement-strategy --placement-strategy type="binpack",field="memory" \
--count 8

Cluster constraints

In the first step, eliminate all the instances that don’t have the required resources based on what you defined either in the JSON task definition or what you provided overrides for to RunTask.

Not enough CPU? Not enough memory? A port is needed, but it is already in use on that instance? Then the instance is eliminated from the set of valid candidates.

Task Placement Cluster Constraints

aws ecs run-task --task-definition nouvelleApp

Placement constraints

In the second step, keep only the instances that satisfy the attribute or task group constraints. Yes, this means that you can indicate what instance to use for a task (for example, to make sure that CPU-intensive jobs are scheduled on the right type of instance, or in which Availability Zone).

You can also create any custom tags of your choosing. The green tasks on the green instances, the blue tasks on the blue instances! You can also use the Cluster Query Language to write expressions to check for multiple attributes. In the next section, I cover how to write and use the attributes and expressions.

Placement Constraints

--placement-constraints type="memberOf",expression="attribute:ecs.instance-type == t2.micro"

Placement strategies

In the third step, filter on the following supported task placement strategies:

  • random
  • binpack
  • spread

By default, tasks are randomly placed with RunTask or spread across Availability Zones with CreateService. Spread is typically used to achieve high availability by making sure that multiple copies of a task are scheduled across multiple instances based on attributes such as Availability Zones.

Conversely, binpack places tasks together to be as cost-efficient as possible. Later in this post, you’ll see how these placement strategies work, as well as how to chain them together and why you may want to do so.

Task Placement Binpack

--placement-strategy type="binpack",field="memory"

Task copies

This isn’t part of the filter, but instead, the count flag is used to indicate how many copies (n) of a given task to run. Effectively, it tells ECS to re-run this workflow n times. By default, the count is set to 1, so run-task is executed one time. For services, the desired-count flag is used.

--count 8

Attributes, task groups, and expressions

For task placement, you can use instance fields, such as attributes, as well as task groups. These can be used in expressions for task placement constraints, or instance fields can be used standalone for task placement strategies. Here’s a quick overview of attributes, task groups, and expressions before you go any further.

Instance: Fields

Because you are using these fields with respect to instances in task placement, the instance: preface is optional and can be used either of the following ways with a field name or an attribute.

instance:<field>
<field>

Field names

The currently supported field names are as follows:

ec2InstanceId
agentConnected

Attributes

There are also instance attributes, which are prefaced with attribute. Again, instance: is optional:

attribute:<attribute-name>

Built-in attributes

The following are some of the provided attributes:

ecs.ami-id
ecs.availability-zone
ecs.instance-type
ecs.os-type
ecs.subnet-id
ecs.vpc-id

Custom attributes

Well, what if you don’t see an attribute that you want? This is where custom attributes come in handy! Want to differentiate between test and prod? What about blue versus green?

aws ecs put-attributes \
--attributes name=color,value=blue,targetId=<your-container-instance-arn>

Task groups

In addition to placing tasks based on attributes, you can use task groups. Every task is assigned a group ID that you can reference in placement. For both tasks and services, a default ID is given, or you can choose your own. Perhaps you want to run version 2 of a service but only on instances with version 1.

task:group

Expressions

Alright, so you have some attributes and task groups… now what? Well, AWS created the Cluster Query Language to make it easy to create expressions for task placement constraints. These attributes and task groups are used with the available comparison operators, which may look familiar if you’ve used Boolean operators before. Some of these operators can be written in multiple ways, such as “!” or “not”.

For instance, to create an expression using a single attribute to select only t2.micro instances, use the ecs.instance-type attribute and the string equality comparator as follows:

attribute:ecs.instance-type == t2.micro

For t2.micro and t2.nano instances, you have a few options. You could use the same syntax as earlier with the or comparator:

attribute:ecs.instance-type == t2.micro or attribute:ecs.instance-type == t2.nano

Another way is to use the in comparator with an argument list:

attribute:ecs.instance-type in [t2.micro, t2.nano]

To include all t2 instances, use a wildcard and the pattern match operator instead of listing out each one:

attribute:ecs.instance-type =~ t2.*

Task group comparisons work the same way. The following snippet selects any instance upon which the task group “database” is running:

task:group == database

To select only task groups that are not “database,” combine expressions:

not(task:group == database)

You can use these expressions to filter your instances:

aws ecs list-container-instances \
--filter "attribute:ecs.instance-type != t2.micro"
aws ecs list-container-instances \
--filter "attribute:color == blue"
aws ecs list-container-instances \
--filter "task:group == database"

These expressions and attributes, respectively, are also used for task placement constraints and strategies, which I cover in the next few sections.

Constraints

Now look at placement constraints. When determining task placement, there may be certain EC2 instances to include or exclude from running containers. For example, you may want to place tasks only on GPU types.

Task placement constraints let you define where your containers should run across your cluster. ECS currently supports two types of placement constraints: distinctInstance and memberOf. By default, ECS spreads tasks across Availability Zones and instances.

  "placementConstraints": [ 
      { 
         "expression": "string",
         "type": "string"
      }
   ],

Distinct Instance

Distinct InstanceThe distinctInstance constraint makes it possible to ensure that every container is started on a unique instance in your cluster. The distinctInstance constraint never places multiple copies of a task on a single instance, even if you request more running tasks than available instances.

For example if you decide to place five copies of a task, each time it filters out the instances that are already running the task.

aws ecs run-task --task-definition nouvelleApp \
--count 5 --placement-constraints type="distinctInstance"

Member of

Member of t2-micro The memberOf constraint describes a set of instances on which your tasks should run. It is for anything you could define as an attribute or task. It also takes in an expression of attributes written in the Cluster Query Language.

For example, if you have a small application and just want it to run on t2.micro instances:

aws ecs run-task --task-definition nouvelleApp \
--count 5 \
--placement-constraints 
type="memberOf",expression="attribute:ecs.instance-type == t2.micro"

You can create expressions using the Cluster Query Language to check for multiple attributes. Here’s how you can weed out all instances in the us-west-2c Availability Zone as well as instances that aren’t of type t2.nano or t2.micro:

aws ecs run-task --task-definition nouvelleApp \
--count 5 \
--placement-constraints type="memberOf",expression="attribute:ecs.availability-zone != us-west-2c and (attribute:ecs.instance-type == t2.nano or attribute:ecs.instance-type == t2.micro)"

Member of affinity

You can also use constraints to place all tasks with the same task group on the same instance (affinity):

aws ecs run-task --task-definition nouvelleApp \
--count 5 --group webserver \
--placement-constraints type=memberOf,expression="task:group == webserver"

Or you can ensure that instances never have more than one task in the same group (anti-affinity):

aws ecs run-task –task-definition nouvelleApp –count 5 –group webserver –placement-constraints type=memberOf,expression=”not(task:group == webserver)”

Strategies

Now look at placement strategies. Placement strategies are used to identify an instance that meets a specific strategy. ECS supports three task placement strategies:

  • random
  • binpack
  • spread

Random is how RunTask places tasks by default and is fairly straightforward (it doesn’t require further parameters). The two other strategies, binpack and spread, take opposite actions. Binpack places tasks on as few instances as possible, helping to optimize resource utilization, while spread places tasks evenly across your cluster to help maximize availability. By default, ECS uses spread with the ecs.availability-zone attribute to place tasks.

   "placementStrategy": [ 
      { 
         "field": "string",
         "type": "string"
      }
   ],

Random

Placement Random

 Random places tasks on instances at random. This still honors the other constraints that you specified, implicitly or explicitly. Specifically, it still makes sure that tasks are scheduled on instances with enough resources to run them.

aws ecs run-task --task-definition nouvelleApp \
--count 5 \
--placement-strategy type="random"

Bin packing

Placement Binpack

The binpack strategy tries to fit your workloads in as few instances as possible. It gets its name from the bin packing problem where the goal is to fit objects of various sizes in the smallest number of bins. It is well suited to scenarios for minimizing the number of instances in your cluster, perhaps for cost savings, and lends itself well to automatic scaling for elastic workloads, to shut down instances that are not in use.

When you use the binpack strategy, you must also indicate if you are trying to make optimal use of your instances’ CPU or memory. This is done by passing an extra field parameter, which tells the task placement engine which parameter to use to evaluate how “full” your “bins” are. It then chooses the instance with the least available CPU or memory (depending on which you pick). If there are multiple instances with this CPU or memory remaining, it chooses randomly.

aws ecs run-task --task-definition nouvelleApp \
--count 8 --placement-strategy type="binpack",field="cpu"

aws ecs run-task --task-definition nouvelleApp \
--count 8 --placement-strategy type="binpack",field="memory"

Spread

Placement Spread

The spread strategy, contrary to the binpack strategy, tries to put your tasks on as many different instances as possible. It is typically used to achieve high availability and mitigate risks, by making sure that you don’t put all your task-eggs in the same instance-baskets. Spread across Availability Zones, therefore, is the default placement strategy used for services.

When using the spread strategy, you must also indicate a field parameter. It is used to indicate the “bins” that you are considering. The accepted values are instanceID to balance tasks across all instances, host, or attribute key:value pairs such as attribute:ecs.availability-zone to balance tasks across zones. There are several AWS attributes that start with the “ecs” prefix, but you can be creative and create your own attributes.

aws ecs run-task --task-definition nouvelleApp \
--count 8 \
--placement-strategy type="spread",field="attribute:ecs.availability-zone"

Chaining placement strategies

Placement binpack spread

Now that you’ve seen how to use task placement strategies, you can also chain multiple task placement strategies with their respective attributes together. You can have up to five strategy rules per service. Perhaps you want to spread tasks across Availability Zones and binpack:

aws ecs run-task --task-definition nouvelleApp \
--count 8 \
--placement-strategy type="spread",field="attribute:ecs.availability-zone" type="binpack",field="memory"

Use cases

Here are some use cases for task placement so you can see how they can be solved by combining attributes, expressions, constraints, and strategies.

Task creation

Mariya is fairly new to using containers and especially container orchestrators. She wants to try ECS and has a simple application that she first wants to get running on a single node. (Solution: Use the RunTask API.)

aws ecs run-task --task-definition nouvelleApp

Scaling

After trying this, Mariya wants to scale her application to run 10 containers across any available nodes in her cluster. (Solution: This means she needs to run a task using either random or spread placement strategies.)

aws ecs run-task --task-definition nouvelleApp \
--count 10 \
--placement-strategy type="random"

Availability

Mariya then realizes that if she wants her tasks to automatically restart themselves if they fail, or if she wants more than 10 instantiations of her task running, she needs to create a service. (Solution: Create a service.)

aws ecs create-service --task-definition nouvelleApp \
--desiredCount 300 --placement-strategy type="random"

Christopher wants to achieve high availability by distributing his tasks amongst all the instances in his cluster so he minimizes impact if any one host goes down. (Solution: To do this he uses spread placement over host name.)

aws ecs run-task --task-definition nouvelleApp \
--count 9 \
--placement-strategy type="spread",field="host"

Ming-ya wants to run a monitoring container on each instance in her cluster. To help her do this, she creates a service with a high desired count and a distinctInstance placement constraint. The ECS service scheduler ensures that each instance in the cluster runs this task (up to the desired count).

aws ecs create-service --service-name monitoring \
--task-definition monitor \
--desiredCount 500 \
--placement-constraints type="distinctInstance"

Availability and Task Groups

Alex wants to run a fleet of webservers. For performance reasons, they want each webserver to have local access to a caching process that was written by another team. They define their webserver as one task, the caching server as a second task. When they launch their webserver task they uses a placement constraint so that the tasks are only placed on instances that are already hosting the cache task. (Solution: Use placement constraints with a task group.)

aws ecs run-task --task-definition cache \
--group caching --count 9 \
--placement-constraints type="distinctInstance"

aws ecs run-task --task-definition webserver \
--count 9 \
--placement-constraints type="distinctInstance" type="memberOf",expression="task:group == caching"

Availability and resource optimization

Jake wants to achieve high availability, but he has a limited budget and needs to optimize all the resources he uses. (Solution: Take a balanced approach of spreading over availability Availability Zones and binpacking on memory within a zone.)

aws ecs run-task --task-definition nouvelleApp \
--count 9 \
--placement-strategy type="spread",field="attribute:ecs.availability-zone" type="binpack",field="memory"

Instance type selection

Aditya has a GPU workload that they want to run in containers on ECS. He needs to ensure that only GPU-enabled instances are used for this workload. (Solution: Create a service and spread on instance type = G2* or whatever other GPU-enabled instance types are in the cluster)

aws ecs create-service --service-name workload \
--task-definition GPU --desiredCount 30 \
--placement-constraints type="memberOf",expression="attribute:ecs.instance-type =~ g2* or attribute:ecs.instance-type =~ p2*"

Conclusion

You’ve now looked at task placement at a high level, as well as:

  • Attributes, task groups, and expressions
  • Constraints
  • Strategies
  • Example use cases

To dive deeper into any of these aspects, check out Task Placement. Also, feel free to ask any questions!

@tiffanyfayj

 

Building, deploying, and operating containerized applications with AWS Fargate

Post Syndicated from Nathan Taber original https://aws.amazon.com/blogs/compute/building-deploying-and-operating-containerized-applications-with-aws-fargate/

This post was contributed by Jason Umiker, AWS Solutions Architect.

Whether it’s helping facilitate a journey to microservices or deploying existing tools more easily and repeatably, many customers are moving toward containerized infrastructure and workflows. AWS provides many of the services and mechanisms to help you with that.

In this post, I show you how to use Amazon ECS and AWS Fargate, as well as AWS CodeBuild and AWS CodePipeline, for an end-to-end CI/CD container solution.

What is Amazon ECS?

Amazon Elastic Container Service (ECS) helps schedule and orchestrate containers across a fleet of servers. It involves installing an agent on each container host that takes instructions from the ECS control plane and relays them to the local Docker image on each one. ECS makes this easy by providing an optimized Amazon Machine Image (AMI) that launches automatically using the ECS console or CLI and that you can use to launch container hosts yourself.

It is up to you to choose the appropriate instance types, sizes, and quantity for your cluster fleet. You should have the capacity to deploy and scale workloads as well as to spread them across enough failure domains for high availability. Features like Auto Scaling groups help with that.

Also, while AWS provides Amazon Linux and Windows AMIs pre-configured for ECS, you are responsible for ongoing maintenance of the OS, which includes patching and security. Items that require regular patching or updating in this model are the OS, Docker, the ECS agent, and of course the contents of the container images.

Two of the key ECS concepts are Tasks and Services. A task is one or more containers that are to be scheduled together by ECS. A service is like an Auto Scaling group for tasks. It defines the quantity of tasks to run across the cluster, where they should be running (for example, across multiple Availability Zones), automatically associates them with a load balancer, and horizontally scales based on metrics that you define like CPI or memory utilization.

What is Fargate?

AWS Fargate is a new compute engine for Amazon ECS that runs containers without requiring you to deploy or manage the underlying Amazon EC2 instances. With Fargate, you specify an image to deploy and the amount of CPU and memory it requires. Fargate handles the updating and securing of the underlying Linux OS, Docker daemon, and ECS agent as well as all the infrastructure capacity management and scaling.

How to use Fargate?

Fargate is exposed as a launch type for ECS. It uses an ECS task and service definition that is similar to the traditional EC2 launch mode, with a few minor differences. It is easy to move tasks and services back and forth between launch types. The differences include:

  • Using the awsvpc network mode
  • Specifying the CPU and memory requirements for the task in the definition

The best way to learn how to use Fargate is to walk through the process and see it in action.

Walkthrough: Deploying a service with Fargate in the console

At the time of publication, Fargate for ECS is available in the N. Virginia, Ohio, Oregon, and Ireland AWS regions. This walkthrough works in any AWS region where Fargate is available.

If you’d prefer to use a CloudFormation template, this one covers Steps 1-4. After launching this template you can skip ahead to Explore Running Service after Step 4.

Step 1 – Create an ECS cluster

An ECS cluster is a logical construct for running groups of containers known as tasks. Clusters can also be used to segregate different environments or teams from each other. In the traditional EC2 launch mode, there are specific EC2 instances associated with and managed by each ECS cluster, but this is transparent to the customer with Fargate.

  1. Open the ECS console and ensure that Fargate is available in the selected Region (for example, N. Virginia).
  2. Choose Clusters, Create Cluster.
  3. Choose Networking only, Next step.
  4. For Cluster name, enter “Fargate”. If you don’t already have a VPC to use, select the Create VPC check box and accept the defaults as well. Choose Create.

Step 2 – Create a task definition, CloudWatch log group, and task execution role

A task is a collection of one or more containers that is the smallest deployable unit of your application. A task definition is a JSON document that serves as the blueprint for ECS to know how to deploy and run your tasks.

The console makes it easier to create this definition by exposing all the parameters graphically. In addition, the console creates two dependencies:

  • The Amazon CloudWatch log group to store the aggregated logs from the task
  • The task execution IAM role that gives Fargate the permissions to run the task
  1. In the left navigation pane, choose Task Definitions, Create new task definition.
  2. Under Select launch type compatibility, choose FARGATE, Next step.
  3. For Task Definition Name, enter NGINX.
  4. If you had an IAM role for your task, you would enter it in Task Role but you don’t need one for this example.
  5. The Network Mode is automatically set to awsvpc for Fargate
  6. Under Task size, for Task memory, choose 0.5 GB. For Task CPU, enter 0.25.
  7. Choose Add container.
  8. For Container name, enter NGINX.
  9. For Image, put nginx:1.13.9-alpine.
  10. For Port mappings type 80 into Container port.
  11. Choose Add, Create.

Step 3 – Create an Application Load Balancer

Sending incoming traffic through a load balancer is often a key piece of making an application both scalable and highly available. It can balance the traffic between multiple tasks, as well as ensure that traffic is only sent to healthy tasks. You can have the service manage the addition or removal of tasks from an Application Load Balancer as they come and go but that must be specified when the service is created. It’s a dependency that you create first.

  1. Open the EC2 console.
  2. In the left navigation pane, choose Load Balancers, Create Load Balancer.
  3. Under Application Load Balancer, choose Create.
  4. For Name, put NGINX.
  5. Choose the appropriate VPC (10.0.0.0/16 if you let ECS create if for you).
  6. For Availability Zones, select both and choose Next: Configure Security Settings.
  7. Choose Next: Configure Security Groups.
  8. For Assign a security group, choose Create a new security group. Choose Next: Configure Routing.
  9. For Name, enter NGINX. For Target type, choose ip.
  10. Choose Next: Register Targets, Next: Review, Create.
  11. Select the new load balancer and note its DNS name (this is the public address for the service).

Step 4 – Create an ECS service using Fargate

A service in ECS using Fargate serves a similar purpose to an Auto Scaling group in EC2. It ensures that the needed number of tasks are running both for scaling as well as spreading the tasks over multiple Availability Zones for high availability. A service creates and destroys tasks as part of its role and can optionally add or remove them from an Application Load Balancer as targets as it does so.

  1. Open the ECS console and ensure that that Fargate is available in the selected Region (for example, N. Virginia).
  2. In the left navigation pane, choose Task Definitions.
  3. Select the NGINX task definition that you created and choose Actions, Create Service.
  4. For Launch Type, select Fargate.
  5. For Service name, enter NGINX.
  6. For Number of tasks, enter 1.
  7. Choose Next step.
  8. Under Subnets, choose both of the options.
  9. For Load balancer type, choose Application Load Balancer. It should then default to the NGINX version that you created earlier.
  10. Choose Add to load balancer.
  11. For Target group name, choose NGINX.
  12. Under DNS records for service discovery, for TTL, enter 60.
  13. Click Next step, Next step, and Create Service.

Explore the running service

At this point, you have a running NGINX service using Fargate. You can now explore what you have running and how it works. You can also ask it to scale up to two tasks across two Availability Zones in the console.

Go into the service and see details about the associated load balancer, tasks, events, metrics, and logs:

Scale the service from one task to multiple tasks:

  • Choose Update.
  • For Number of tasks, enter 2.
  • Choose Next step, Next step, Next step then Update Service.
  • Watch the event that is logged and the new additional task both appear.

On the service Details tab, open the NGINX Target Group Name link and see the IP address registered targets spread across the two zones.

Go to the DNS name for the Application Load Balancer in your browser and see the default NGINX page. Get the value from the Load Balancers dashboard in the EC2 console.

Walkthrough: Adding a CI/CD pipeline to your service

Now, I’m going to show you how to set up a CI/CD pipeline around this service. It watches a GitHub repo for changes and rebuilds the container with CodeBuild based on the buildspec.yml file and Dockerfile in the repo. If that build is successful, it then updates your Fargate service to deploy the new image.

If you’d prefer to use a CloudFormation Template, this one covers the creation of the dependencies so that the console will pre-fill these (CodeBuild Project and IAM Roles) during the creation of the CodePipeline in the steps below.

Step 1 – Create an ECR repository for the rebuilt container image

An ECR repository is a place to store your container images in a secure and reliable manner. Scaling and self-healing of Fargate tasks requires these images to be always available to be pulled when required. This is an important part of a container platform.

  1. Open the ECS console and ensure that that Fargate is available in the selected Region (for example N. Virginia).
  2. In the left navigation pane, under Amazon ECR, choose Repositories, Get started.
  3. For Repository name, put NGINX and choose Next step.

Step 2 – Fork the nginx-codebuild example into your own GitHub account

I have created an example project that takes the Dockerfile and config files for the official NGINX Docker Hub image and adds a buildspec.yml file to tell CodeBuild how to build the container and push it to your new ECR registry on completion. You can fork it into your own GitHub account for this CI/CD demo.

  1. Go to https://github.com/jasonumiker/nginx-codebuild.
  2. In the upper right corner, choose Fork.

Step 3 – Create the pipeline and associated IAM roles

You have two complementary AWS services for building a CI/CD pipeline for your containers. CodeBuild executes the build jobs and CodePipeline kicks off those builds when it notices that the source GitHub or CodeCommit repo changes. If successful, CodePipeline then deploys the new container image to Fargate.

The CodePipeline console can create the associated CodeBuild project, in addition to other dependencies such as the required IAM roles.

  1. Open the CodePipeline console and ensure that that Fargate is available in the selected Region (for example, N. Virginia).
  2. Choose Get started.
  3. For Pipeline name, enter NGINX and choose Next step.
  4. For Source provider, choose GitHub.
  5. Choose Connect to GitHub and log in.
    • For Repository, choose your forked nginx-codebuild repo. For Branch, enter master. Choose Next step.
  6. For Build provider, enter AWS CodeBuild.
  7. Select Create a new build project.
  8. For Project name, enter NGINX.
  9. For Operating system, choose Ubuntu. For Runtime, choose Docker. For Version, select the latest version.
  10. Expand Advanced and set the following environment variables:
    • AWS_ACCOUNT_ID with a value of the account number
    • IMAGE_REPO_NAME with a value of NGINX (or whatever ECR name that you used)
  11. Choose Save build project, Next step.
  12. For Deployment provider, choose Amazon ECS.
  13. For Cluster name, enter Fargate.
  14. For Service name, choose NGINX.
  15. For Image filename, enter images.json.
  16. Choose Next step.
  17. Choose Create role, Allow, Next step, and then choose Create pipeline.
  18. Open the IAM console and ensure that that Fargate is available in the selected Region (for example, N. Virginia).
  19. In the left navigation pane, choose Roles.
  20. Choose the code-build-nginx-service-role that was just created and choose Attach policy.
  21. For Policy type, choose AmazonEC2ContainerRegistryPowerUser and choose Attach policy.

Step 4 – Start the pipeline

You now have CodePipeline watching the GitHub repo for changes. It kicks off a CodeBuild build job on a change and, if the build is successful, creates a new deployment of the Fargate service with the new image.

Make a change to the source repo (even just adding a new dummy file) and then commit it and push it to master on your GitHub fork. This automatically kicks off the pipeline to build and deploy the change.

Conclusion

As you’ve seen, Fargate is fast and easy to set up, integrates well with the rest of the AWS platform, and saves you from much of the heavy lifting of running containers reliably at scale.

While it is useful to go through creating things in the console to understand them better we suggest automating them with infrastructure-as-code patterns via things like our CloudFormation to ensure that they are repeatable, and any changes can be managed. There are some example templates to help you get started in this post.

In addition, adding things like unit and integration testing, blue/green and/or manual approval gates into CodePipeline are often a good idea before deploying patterns like this to production in many organizations. Some additional examples to look at next include:

EC2 Instance Update – M5 Instances with Local NVMe Storage (M5d)

Post Syndicated from Jeff Barr original https://aws.amazon.com/blogs/aws/ec2-instance-update-m5-instances-with-local-nvme-storage-m5d/

Earlier this month we launched the C5 Instances with Local NVMe Storage and I told you that we would be doing the same for additional instance types in the near future!

Today we are introducing M5 instances equipped with local NVMe storage. Available for immediate use in 5 regions, these instances are a great fit for workloads that require a balance of compute and memory resources. Here are the specs:

Instance NamevCPUsRAMLocal StorageEBS-Optimized BandwidthNetwork Bandwidth
m5d.large28 GiB1 x 75 GB NVMe SSDUp to 2.120 GbpsUp to 10 Gbps
m5d.xlarge416 GiB1 x 150 GB NVMe SSDUp to 2.120 GbpsUp to 10 Gbps
m5d.2xlarge832 GiB1 x 300 GB NVMe SSDUp to 2.120 GbpsUp to 10 Gbps
m5d.4xlarge1664 GiB1 x 600 GB NVMe SSD2.210 GbpsUp to 10 Gbps
m5d.12xlarge48192 GiB2 x 900 GB NVMe SSD5.0 Gbps10 Gbps
m5d.24xlarge96384 GiB4 x 900 GB NVMe SSD10.0 Gbps25 Gbps

The M5d instances are powered by Custom Intel® Xeon® Platinum 8175M series processors running at 2.5 GHz, including support for AVX-512.

You can use any AMI that includes drivers for the Elastic Network Adapter (ENA) and NVMe; this includes the latest Amazon Linux, Microsoft Windows (Server 2008 R2, Server 2012, Server 2012 R2 and Server 2016), Ubuntu, RHEL, SUSE, and CentOS AMIs.

Here are a couple of things to keep in mind about the local NVMe storage on the M5d instances:

Naming – You don’t have to specify a block device mapping in your AMI or during the instance launch; the local storage will show up as one or more devices (/dev/nvme*1 on Linux) after the guest operating system has booted.

Encryption – Each local NVMe device is hardware encrypted using the XTS-AES-256 block cipher and a unique key. Each key is destroyed when the instance is stopped or terminated.

Lifetime – Local NVMe devices have the same lifetime as the instance they are attached to, and do not stick around after the instance has been stopped or terminated.

Available Now
M5d instances are available in On-Demand, Reserved Instance, and Spot form in the US East (N. Virginia), US West (Oregon), EU (Ireland), US East (Ohio), and Canada (Central) Regions. Prices vary by Region, and are just a bit higher than for the equivalent M5 instances.

— Jeff;

 

EC2 Instance Update – C5 Instances with Local NVMe Storage (C5d)

Post Syndicated from Jeff Barr original https://aws.amazon.com/blogs/aws/ec2-instance-update-c5-instances-with-local-nvme-storage-c5d/

As you can see from my EC2 Instance History post, we add new instance types on a regular and frequent basis. Driven by increasingly powerful processors and designed to address an ever-widening set of use cases, the size and diversity of this list reflects the equally diverse group of EC2 customers!

Near the bottom of that list you will find the new compute-intensive C5 instances. With a 25% to 50% improvement in price-performance over the C4 instances, the C5 instances are designed for applications like batch and log processing, distributed and or real-time analytics, high-performance computing (HPC), ad serving, highly scalable multiplayer gaming, and video encoding. Some of these applications can benefit from access to high-speed, ultra-low latency local storage. For example, video encoding, image manipulation, and other forms of media processing often necessitates large amounts of I/O to temporary storage. While the input and output files are valuable assets and are typically stored as Amazon Simple Storage Service (S3) objects, the intermediate files are expendable. Similarly, batch and log processing runs in a race-to-idle model, flushing volatile data to disk as fast as possible in order to make full use of compute resources.

New C5d Instances with Local Storage
In order to meet this need, we are introducing C5 instances equipped with local NVMe storage. Available for immediate use in 5 regions, these instances are a great fit for the applications that I described above, as well as others that you will undoubtedly dream up! Here are the specs:

Instance NamevCPUsRAMLocal StorageEBS BandwidthNetwork Bandwidth
c5d.large24 GiB1 x 50 GB NVMe SSDUp to 2.25 GbpsUp to 10 Gbps
c5d.xlarge48 GiB1 x 100 GB NVMe SSDUp to 2.25 GbpsUp to 10 Gbps
c5d.2xlarge816 GiB1 x 225 GB NVMe SSDUp to 2.25 GbpsUp to 10 Gbps
c5d.4xlarge1632 GiB1 x 450 GB NVMe SSD2.25 GbpsUp to 10 Gbps
c5d.9xlarge3672 GiB1 x 900 GB NVMe SSD4.5 Gbps10 Gbps
c5d.18xlarge72144 GiB2 x 900 GB NVMe SSD9 Gbps25 Gbps

Other than the addition of local storage, the C5 and C5d share the same specs. Both are powered by 3.0 GHz Intel Xeon Platinum 8000-series processors, optimized for EC2 and with full control over C-states on the two largest sizes, giving you the ability to run two cores at up to 3.5 GHz using Intel Turbo Boost Technology.

You can use any AMI that includes drivers for the Elastic Network Adapter (ENA) and NVMe; this includes the latest Amazon Linux, Microsoft Windows (Server 2008 R2, Server 2012, Server 2012 R2 and Server 2016), Ubuntu, RHEL, SUSE, and CentOS AMIs.

Here are a couple of things to keep in mind about the local NVMe storage:

Naming – You don’t have to specify a block device mapping in your AMI or during the instance launch; the local storage will show up as one or more devices (/dev/nvme*1 on Linux) after the guest operating system has booted.

Encryption – Each local NVMe device is hardware encrypted using the XTS-AES-256 block cipher and a unique key. Each key is destroyed when the instance is stopped or terminated.

Lifetime – Local NVMe devices have the same lifetime as the instance they are attached to, and do not stick around after the instance has been stopped or terminated.

Available Now
C5d instances are available in On-Demand, Reserved Instance, and Spot form in the US East (N. Virginia), US West (Oregon), EU (Ireland), US East (Ohio), and Canada (Central) Regions. Prices vary by Region, and are just a bit higher than for the equivalent C5 instances.

— Jeff;

PS – We will be adding local NVMe storage to other EC2 instance types in the months to come, so stay tuned!

Introducing the AWS Machine Learning Competency for Consulting Partners

Post Syndicated from Randall Hunt original https://aws.amazon.com/blogs/aws/introducing-the-aws-machine-learning-competency-for-consulting-partners/

Today I’m excited to announce a new Machine Learning Competency for Consulting Partners in the Amazon Partner Network (APN). This AWS Competency program allows APN Consulting Partners to demonstrate a deep expertise in machine learning on AWS by providing solutions that enable machine learning and data science workflows for their customers. This new AWS Competency is in addition to the Machine Learning comptency for our APN Technology Partners, that we launched at the re:Invent 2017 partner summit.

These APN Consulting Partners help organizations solve their machine learning and data challenges through:

  • Providing data services that help data scientists and machine learning practitioners prepare their enterprise data for training.
  • Platform solutions that provide data scientists and machine learning practitioners with tools to take their data, train models, and make predictions on new data.
  • SaaS and API solutions to enable predictive capabilities within customer applications.

Why work with an AWS Machine Learning Competency Partner?

The AWS Competency Program helps customers find the most qualified partners with deep expertise. AWS Machine Learning Competency Partners undergo a strict validation of their capabilities to demonstrate technical proficiency and proven customer success with AWS machine learning tools.

If you’re an AWS customer interested in machine learning workloads on AWS, check out our AWS Machine Learning launch partners below:

 

Interested in becoming an AWS Machine Learning Competency Partner?

APN Partners with experience in Machine Learning can learn more about becoming an AWS Machine Learning Competency Partner here. To learn more about the benefits of joining the AWS Partner Network, see our APN Partner website.

Thanks to the AWS Partner Team for their help with this post!
– Randall

What’s new in HiveMQ 3.4

Post Syndicated from The HiveMQ Team original https://www.hivemq.com/whats-new-in-hivemq-3-4

We are pleased to announce the release of HiveMQ 3.4. This version of HiveMQ is the most resilient and advanced version of HiveMQ ever. The main focus in this release was directed towards addressing the needs for the most ambitious MQTT deployments in the world for maximum performance and resilience for millions of concurrent MQTT clients. Of course, deployments of all sizes can profit from the improvements in the latest and greatest HiveMQ.

This version is a drop-in replacement for HiveMQ 3.3 and of course supports rolling upgrades with zero-downtime.

HiveMQ 3.4 brings many features that your users, administrators and plugin developers are going to love. These are the highlights:

 

New HiveMQ 3.4 features at a glance

Cluster

WebUI

Plugin System

Plugin System

Operations

Other notable improvements

Cluster

HiveMQ 3.4 brings various improvements in terms of scalability, availability, resilience and observability for the cluster mechanism. Many of the new features remain under the hood, but several additions stand out:

Cluster Overload Protection

The new version has a first-of-its-kind Cluster Overload Protection. The whole cluster is able to spot MQTT clients that cause overload on nodes or the cluster as a whole and protects itself from the overload. This mechanism also protects the deployment from cascading failures due to slow or failing underlying hardware (as sometimes seen on cloud providers). This feature is enabled by default and you can learn more about the mechanism in our documentation.

Dynamic Replicates

HiveMQ’s sophisticated cluster mechanism is able to scale in a linear fashion due to extremely efficient and true data distribution mechanics based on a configured replication factor. The most important aspect of every cluster is availability, which is achieved by having eventual consistency functions in place for edge cases. The 3.4 version adds dynamic replicates to the cluster so even the most challenging edge cases involving network splits don’t lead to the sacrifice of consistency for the most important MQTT operations.

Node Stress Level Metrics

All MQTT cluster nodes are now aware of their own stress level and the stress levels of other cluster members. While all stress mitigation is handled internally by HiveMQ, experienced operators may want to monitor the individual node’s stress level (e.g with Grafana) in order to start investigating what caused the increase of load.

WebUI

Operators worldwide love the HiveMQ WebUI introduced with HiveMQ 3.3. We gathered all the fantastic feedback from our users and polished the WebUI, so it’s even more useful for day-to-day broker operations and remote debugging of MQTT clients. The most important changes and additions are:

Trace Recording Download

The unique Trace Recordings functionality is without doubt a lifesaver when the behavior of individual MQTT clients needs further investigation as all interactions with the broker can be traced — at runtime and at scale! Huge production deployments may accumulate multiple gigabytes of trace recordings. HiveMQ now offers a convenient way to collect all trace recordings from all nodes, zips them and allows the download via a simple button on the WebUI. Remote debugging was never easier!

Additional Client Detail Information in WebUI

The mission of the HiveMQ WebUI is to provide easy insights to the whole production MQTT cluster for operators and administrators. Individual MQTT client investigations are a piece of cake, as all available information about clients can be viewed in detail. We further added the ability to view the restrictions a concrete client has:

  • Maximum Inflight Queue Size
  • Client Offline Queue Messages Size
  • Client Offline Message Drop Strategy

Session Invalidation

MQTT persistent sessions are one of the outstanding features of the MQTT protocol specification. Sessions which do not expire but are never reused unnecessarily consume disk space and memory. Administrators can now invalidate individual session directly in the HiveMQ WebUI for client sessions, which can be deleted safely. HiveMQ 3.4 will take care and release the resources on all cluster nodes after a session was invalidated

Web UI Polishing

Most texts on the WebUI were revisited and are now clearer and crisper. The help texts also received a major overhaul and should now be more, well, helpful. In addition, many small improvements were added, which are most of the time invisible but are here to help when you need them most. For example, the WebUI now displays a warning if cluster nodes with old versions are in the cluster (which may happen if a rolling upgrade was not finished properly)

Plugin System

One of the most popular features of HiveMQ is the extensive Plugin System, which virtually enables the integration of HiveMQ to any system and allows hooking into all aspects of the MQTT lifecycle. We listened to the feedback and are pleased to announce many improvements, big and small, for the Plugin System:

Client Session Time-to-live for individual clients

HiveMQ 3.3 offered a global configuration for setting the Time-To-Live for MQTT sessions. With the advent of HiveMQ 3.4, users can now programmatically set Time-To-Live values for individual MQTT clients and can discard a MQTT session immediately.

Individual Inflight Queues

While the Inflight Queue configuration is typically sufficient in the HiveMQ default configuration, there are some use cases that require the adjustment of this configuration. It’s now possible to change the Inflight Queue size for individual clients via the Plugin System.
 
 

Plugin Service Overload Protection

The HiveMQ Plugin System is a power-user tool and it’s possible to do unbelievably useful modifications as well as putting major stress on the system as a whole if the programmer is not careful. In order to protect the HiveMQ instances from accidental overload, a Plugin Service Overload Protection can be configured. This rate limits the Plugin Service usage and gives feedback to the application programmer in case the rate limit is exceeded. This feature is disabled by default but we strongly recommend updating your plugins to profit from this feature.

Session Attribute Store putIfNewer

This is one of the small bits you almost never need but when you do, you’re ecstatic for being able to use it. The Session Attribute Store now offers methods to put values, if the values you want to put are newer or fresher than the values already written. This is extremely useful, if multiple cluster nodes want to write to the Session Attribute Store simultaneously, as this guarantees that outdated values can no longer overwrite newer values.
 
 
 
 

Disconnection Timestamp for OnDisconnectCallback

As the OnDisconnectCallback is executed asynchronously, the client might already be gone when the callback is executed. It’s now easy to obtain the exact timestamp when a MQTT client disconnected, even if the callback is executed later on. This feature might be very interesting for many plugin developers in conjunction with the Session Attribute Store putIfNewer functionality.

Operations

We ❤️ Operators and we strive to provide all the tools needed for operating and administrating a MQTT broker cluster at scale in any environment. A key strategy for successful operations of any system is monitoring. We added some interesting new metrics you might find useful.

System Metrics

In addition to JVM Metrics, HiveMQ now also gathers Operating System Metrics for Linux Systems. So HiveMQ is able to see for itself how the operating system views the process, including native memory, the real CPU usage, and open file usage. These metrics are particularly useful, if you don’t have a monitoring agent for Linux systems setup. All metrics can be found here.

Client Disconnection Metrics

The reality of many MQTT scenarios is that not all clients are able to disconnect gracefully by sending MQTT DISCONNECT messages. HiveMQ now also exposes metrics about clients that disconnected by closing the TCP connection instead of sending a DISCONNECT packet first. This is especially useful for monitoring, if you regularly deal with clients that don’t have a stable connection to the MQTT brokers.

 

JMX enabled by default

JMX, the Java Monitoring Extension, is now enabled by default. Many HiveMQ operators use Application Performance Monitoring tools, which are able to hook into the metrics via JMX or use plain JMX for on-the-fly debugging. While we recommend to use official off-the-shelf plugins for monitoring, it’s now easier than ever to just use JMX if other solutions are not available to you.

Other notable improvements

The 3.4 release of HiveMQ is full of hidden gems and improvements. While it would be too much to highlight all small improvements, these notable changes stand out and contribute to the best HiveMQ release ever.

Topic Level Distribution Configuration

Our recommendation for all huge deployments with millions of devices is: Start with separate topic prefixes by bringing the dynamic topic parts directly to the beginning. The reality is that many customers have topics that are constructed like the following: “devices/{deviceId}/status”. So what happens is that all topics in this example start with a common prefix, “devices”, which is the first topic level. Unfortunately the first topic level doesn’t include a dynamic topic part. In order to guarantee the best scalability of the cluster and the best performance of the topic tree, customers can now configure how many topic levels are used for distribution. In the example outlined here, a topic level distribution of 2 would be perfect and guarantees the best scalability.

Mass disconnect performance improvements

Mass disconnections of MQTT clients can happen. This might be the case when e.g. a load balancer in front of the MQTT broker cluster drops the connections or if a mobile carrier experiences connectivity problems. Prior to HiveMQ 3.4, mass disconnect events caused stress on the cluster. Mass disconnect events are now massively optimized and even tens of millions of connection losses at the same time won’t bring the cluster into stress situations.

 
 
 
 
 
 

Replication Performance Improvements

Due to the distributed nature of a HiveMQ, data needs to be replicated across the cluster in certain events, e.g. when cluster topology changes occur. There are various internal improvements in HiveMQ version 3.4, which increase the replication performance significantly. Our engineers put special love into the replication of Queued Messages, which is now faster than ever, even for multiple millions of Queued Messages that need to be transferred across the cluster.

Updated Native SSL Libraries

The Native SSL Integration of HiveMQ was updated to the newest BoringSSL version. This results in better performance and increased security. In case you’re using SSL and you are not yet using the native SSL integration, we strongly recommend to give it a try, more than 40% performance improvement can be observed for most deployments.

 
 

Improvements for Java 9

While Java 9 was already supported for older HiveMQ versions, HiveMQ 3.4 has full-blown Java 9 support. The minimum Java version still remains Java 7, although we strongly recommend to use Java 8 or newer for the best performance of HiveMQ.

EC2 Fleet – Manage Thousands of On-Demand and Spot Instances with One Request

Post Syndicated from Jeff Barr original https://aws.amazon.com/blogs/aws/ec2-fleet-manage-thousands-of-on-demand-and-spot-instances-with-one-request/

EC2 Spot Fleets are really cool. You can launch a fleet of Spot Instances that spans EC2 instance types and Availability Zones without having to write custom code to discover capacity or monitor prices. You can set the target capacity (the size of the fleet) in units that are meaningful to your application and have Spot Fleet create and then maintain the fleet on your behalf. Our customers are creating Spot Fleets of all sizes. For example, one financial service customer runs Monte Carlo simulations across 10 different EC2 instance types. They routinely make requests for hundreds of thousands of vCPUs and count on Spot Fleet to give them access to massive amounts of capacity at the best possible price.

EC2 Fleet
Today we are extending and generalizing the set-it-and-forget-it model that we pioneered in Spot Fleet with EC2 Fleet, a new building block that gives you the ability to create fleets that are composed of a combination of EC2 On-Demand, Reserved, and Spot Instances with a single API call. You tell us what you need, capacity and instance-wise, and we’ll handle all the heavy lifting. We will launch, manage, monitor and scale instances as needed, without the need for scaffolding code.

You can specify the capacity of your fleet in terms of instances, vCPUs, or application-oriented units, and also indicate how much of the capacity should be fulfilled by Spot Instances. The application-oriented units allow you to specify the relative power of each EC2 instance type in a way that directly maps to the needs of your application. All three capacity specification options (instances, vCPUs, and application-oriented units) are known as weights.

I think you’ll find a number ways this feature makes managing a fleet of instances easier, and believe that you will also appreciate the team’s near-term feature roadmap of interest (more on that in a bit).

Using EC2 Fleet
There are a number of ways that you can use this feature, whether you’re running a stateless web service, a big data cluster or a continuous integration pipeline. Today I’m going to describe how you can use EC2 Fleet for genomic processing, but this is similar to workloads like risk analysis, log processing or image rendering. Modern DNA sequencers can produce multiple terabytes of raw data each day, to process that data into meaningful information in a timely fashion you need lots of processing power. I’ll be showing you how to deploy a “grid” of worker nodes that can quickly crunch through secondary analysis tasks in parallel.

Projects in genomics can use the elasticity EC2 provides to experiment and try out new pipelines on hundreds or even thousands of servers. With EC2 you can access as many cores as you need and only pay for what you use. Prior to today, you would need to use the RunInstances API or an Auto Scaling group for the On-Demand & Reserved Instance portion of your grid. To get the best price performance you’d also create and manage a Spot Fleet or multiple Spot Auto Scaling groups with different instance types if you wanted to add Spot Instances to turbo-boost your secondary analysis. Finally, to automate scaling decisions across multiple APIs and Auto Scaling groups you would need to write Lambda functions that periodically assess your grid’s progress & backlog, as well as current Spot prices – modifying your Auto Scaling Groups and Spot Fleets accordingly.

You can now replace all of this with a single EC2 Fleet, analyzing genomes at scale for as little as $1 per analysis. In my grid, each step in in the pipeline requires 1 vCPU and 4 GiB of memory, a perfect match for M4 and M5 instances with 4 GiB of memory per vCPU. I will create a fleet using M4 and M5 instances with weights that correspond to the number of vCPUs on each instance:

  • m4.16xlarge – 64 vCPUs, weight = 64
  • m5.24xlarge – 96 vCPUs, weight = 96

This is expressed in a template that looks like this:

"Overrides": [
{
  "InstanceType": "m4.16xlarge",
  "WeightedCapacity": 64,
},
{
  "InstanceType": "m5.24xlarge",
  "WeightedCapacity": 96,
},
]

By default, EC2 Fleet will select the most cost effective combination of instance types and Availability Zones (both specified in the template) using the current prices for the Spot Instances and public prices for the On-Demand Instances (if you specify instances for which you have matching RIs, your discounts will apply). The default mode takes weights into account to get the instances that have the lowest price per unit. So for my grid, fleet will find the instance that offers the lowest price per vCPU.

Now I can request capacity in terms of vCPUs, knowing EC2 Fleet will select the lowest cost option using only the instance types I’ve defined as acceptable. Also, I can specify how many vCPUs I want to launch using On-Demand or Reserved Instance capacity and how many vCPUs should be launched using Spot Instance capacity:

"TargetCapacitySpecification": {
	"TotalTargetCapacity": 2880,
	"OnDemandTargetCapacity": 960,
	"SpotTargetCapacity": 1920,
	"DefaultTargetCapacityType": "Spot"
}

The above means that I want a total of 2880 vCPUs, with 960 vCPUs fulfilled using On-Demand and 1920 using Spot. The On-Demand price per vCPU is lower for m5.24xlarge than the On-Demand price per vCPU for m4.16xlarge, so EC2 Fleet will launch 10 m5.24xlarge instances to fulfill 960 vCPUs. Based on current Spot pricing (again, on a per-vCPU basis), EC2 Fleet will choose to launch 30 m4.16xlarge instances or 20 m5.24xlarges, delivering 1920 vCPUs either way.

Putting it all together, I have a single file (fl1.json) that describes my fleet:

    "LaunchTemplateConfigs": [
        {
            "LaunchTemplateSpecification": {
                "LaunchTemplateId": "lt-0e8c754449b27161c",
                "Version": "1"
            }
        "Overrides": [
        {
          "InstanceType": "m4.16xlarge",
          "WeightedCapacity": 64,
        },
        {
          "InstanceType": "m5.24xlarge",
          "WeightedCapacity": 96,
        },
      ]
        }
    ],
    "TargetCapacitySpecification": {
        "TotalTargetCapacity": 2880,
        "OnDemandTargetCapacity": 960,
        "SpotTargetCapacity": 1920,
        "DefaultTargetCapacityType": "Spot"
    }
}

I can launch my fleet with a single command:

$ aws ec2 create-fleet --cli-input-json file://home/ec2-user/fl1.json
{
    "FleetId":"fleet-838cf4e5-fded-4f68-acb5-8c47ee1b248a"
}

My entire fleet is created within seconds and was built using 10 m5.24xlarge On-Demand Instances and 30 m4.16xlarge Spot Instances, since the current Spot price was 1.5¢ per vCPU for m4.16xlarge and 1.6¢ per vCPU for m5.24xlarge.

Now lets imagine my grid has crunched through its backlog and no longer needs the additional Spot Instances. I can then modify the size of my fleet by changing the target capacity in my fleet specification, like this:

{         
    "TotalTargetCapacity": 960,
}

Since 960 was equal to the amount of On-Demand vCPUs I had requested, when I describe my fleet I will see all of my capacity being delivered using On-Demand capacity:

"TargetCapacitySpecification": {
	"TotalTargetCapacity": 960,
	"OnDemandTargetCapacity": 960,
	"SpotTargetCapacity": 0,
	"DefaultTargetCapacityType": "Spot"
}

When I no longer need my fleet I can delete it and terminate the instances in it like this:

$ aws ec2 delete-fleets --fleet-id fleet-838cf4e5-fded-4f68-acb5-8c47ee1b248a \
  --terminate-instances   
{
    "UnsuccessfulFleetDletetions": [],
    "SuccessfulFleetDeletions": [
        {
            "CurrentFleetState": "deleted_terminating",
            "PreviousFleetState": "active",
            "FleetId": "fleet-838cf4e5-fded-4f68-acb5-8c47ee1b248a"
        }
    ]
}

Earlier I described how RI discounts apply when EC2 Fleet launches instances for which you have matching RIs, so you might be wondering how else RI customers benefit from EC2 Fleet. Let’s say that I own regional RIs for M4 instances. In my EC2 Fleet I would remove m5.24xlarge and specify m4.10xlarge and m4.16xlarge. Then when EC2 Fleet creates the grid, it will quickly find M4 capacity across the sizes and AZs I’ve specified, and my RI discounts apply automatically to this usage.

In the Works
We plan to connect EC2 Fleet and EC2 Auto Scaling groups. This will let you create a single fleet that mixed instance types and Spot, Reserved and On-Demand, while also taking advantage of EC2 Auto Scaling features such as health checks and lifecycle hooks. This integration will also bring EC2 Fleet functionality to services such as Amazon ECS, Amazon EKS, and AWS Batch that build on and make use of EC2 Auto Scaling for fleet management.

Available Now
You can create and make use of EC2 Fleets today in all public AWS Regions!

— Jeff;

Secure Build with AWS CodeBuild and LayeredInsight

Post Syndicated from Asif Khan original https://aws.amazon.com/blogs/devops/secure-build-with-aws-codebuild-and-layeredinsight/

This post is written by Asif Awan, Chief Technology Officer of Layered Insight, Subin Mathew – Software Development Manager for AWS CodeBuild, and Asif Khan – Solutions Architect

Enterprises adopt containers because they recognize the benefits: speed, agility, portability, and high compute density. They understand how accelerating application delivery and deployment pipelines makes it possible to rapidly slipstream new features to customers. Although the benefits are indisputable, this acceleration raises concerns about security and corporate compliance with software governance. In this blog post, I provide a solution that shows how Layered Insight, the pioneer and global leader in container-native application protection, can be used with seamless application build and delivery pipelines like those available in AWS CodeBuild to address these concerns.

Layered Insight solutions

Layered Insight enables organizations to unify DevOps and SecOps by providing complete visibility and control of containerized applications. Using the industry’s first embedded security approach, Layered Insight solves the challenges of container performance and protection by providing accurate insight into container images, adaptive analysis of running containers, and automated enforcement of container behavior.

 

AWS CodeBuild

AWS CodeBuild is a fully managed build service that compiles source code, runs tests, and produces software packages that are ready to deploy. With CodeBuild, you don’t need to provision, manage, and scale your own build servers. CodeBuild scales continuously and processes multiple builds concurrently, so your builds are not left waiting in a queue. You can get started quickly by using prepackaged build environments, or you can create custom build environments that use your own build tools.

 

Problem Definition

Security and compliance concerns span the lifecycle of application containers. Common concerns include:

Visibility into the container images. You need to verify the software composition information of the container image to determine whether known vulnerabilities associated with any of the software packages and libraries are included in the container image.

Governance of container images is critical because only certain open source packages/libraries, of specific versions, should be included in the container images. You need support for mechanisms for blacklisting all container images that include a certain version of a software package/library, or only allowing open source software that come with a specific type of license (such as Apache, MIT, GPL, and so on). You need to be able to address challenges such as:

·       Defining the process for image compliance policies at the enterprise, department, and group levels.

·       Preventing the images that fail the compliance checks from being deployed in critical environments, such as staging, pre-prod, and production.

Visibility into running container instances is critical, including:

·       CPU and memory utilization.

·       Security of the build environment.

·       All activities (system, network, storage, and application layer) of the application code running in each container instance.

Protection of running container instances that is:

·       Zero-touch to the developers (not an SDK-based approach).

·       Zero touch to the DevOps team and doesn’t limit the portability of the containerized application.

·       This protection must retain the option to switch to a different container stack or orchestration layer, or even to a different Container as a Service (CaaS ).

·       And it must be a fully automated solution to SecOps, so that the SecOps team doesn’t have to manually analyze and define detailed blacklist and whitelist policies.

 

Solution Details

In AWS CodeCommit, we have three projects:
●     “Democode” is a simple Java application, with one buildspec to build the app into a Docker container (run by build-demo-image CodeBuild project), and another to instrument said container (instrument-image CodeBuild project). The resulting container is stored in ECR repo javatestasjavatest:20180415-layered. This instrumented container is running in AWS Fargate cluster demo-java-appand can be seen in the Layered Insight runtime console as the javatestapplication in us-east-1.
●     aws-codebuild-docker-imagesis a clone of the official aws-codebuild-docker-images repo on GitHub . This CodeCommit project is used by the build-python-builder CodeBuild project to build the python 3.3.6 codebuild image and is stored at the codebuild-python ECR repo. We then manually instructed the Layered Insight console to instrument the image.
●     scan-java-imagecontains just a buildspec.yml file. This file is used by the scan-java-image CodeBuild project to instruct Layered Assessment to perform a vulnerability scan of the javatest container image built previously, and then run the scan results through a compliance policy that states there should be no medium vulnerabilities. This build fails — but in this case that is a success: the scan completes successfully, but compliance fails as there are medium-level issues found in the scan.

This build is performed using the instrumented version of the Python 3.3.6 CodeBuild image, so the activity of the processes running within the build are recorded each time within the LI console.

Build container image

Create or use a CodeCommit project with your application. To build this image and store it in Amazon Elastic Container Registry (Amazon ECR), add a buildspec file to the project and build a container image and create a CodeBuild project.

Scan container image

Once the image is built, create a new buildspec in the same project or a new one that looks similar to below (update ECR URL as necessary):

version: 0.2
phases:
  pre_build:
    commands:
      - echo Pulling down LI Scan API client scripts
      - git clone https://github.com/LayeredInsight/scan-api-example-python.git
      - echo Setting up LI Scan API client
      - cd scan-api-example-python
      - pip install layint_scan_api
      - pip install -r requirements.txt
  build:
    commands:
      - echo Scanning container started on `date`
      - IMAGEID=$(./li_add_image --name <aws-region>.amazonaws.com/javatest:20180415)
      - ./li_wait_for_scan -v --imageid $IMAGEID
      - ./li_run_image_compliance -v --imageid $IMAGEID --policyid PB15260f1acb6b2aa5b597e9d22feffb538256a01fbb4e5a95

Add the buildspec file to the git repo, push it, and then build a CodeBuild project using with the instrumented Python 3.3.6 CodeBuild image at <aws-region>.amazonaws.com/codebuild-python:3.3.6-layered. Set the following environment variables in the CodeBuild project:
●     LI_APPLICATIONNAME – name of the build to display
●     LI_LOCATION – location of the build project to display
●     LI_API_KEY – ApiKey:<key-name>:<api-key>
●     LI_API_HOST – location of the Layered Insight API service

Instrument container image

Next, to instrument the new container image:

  1. In the Layered Insight runtime console, ensure that the ECR registry and credentials are defined (click the Setup icon and the ‘+’ sign on the top right of the screen to add a new container registry). Note the name given to the registry in the console, as this needs to be referenced in the li_add_imagecommand in the script, below.
  2. Next, add a new buildspec (with a new name) to the CodeCommit project, such as the one shown below. This code will download the Layered Insight runtime client, and use it to instruct the Layered Insight service to instrument the image that was just built:
    version: 0.2
phases:
pre_build:
commands:
echo Pulling down LI API Runtime client scripts
git clone https://github.com/LayeredInsight/runtime-api-example-python
echo Setting up LI API client
cd runtime-api-example-python
pip install layint-runtime-api
pip install -r requirements.txt
build:
commands:
echo Instrumentation started on `date`
./li_add_image --registry "Javatest ECR" --name IMAGE_NAME:TAG --description "IMAGE DESCRIPTION" --policy "Default Policy" --instrument --wait --verbose
  3. Commit and push the new buildspec file.
  4. Going back to CodeBuild, create a new project, with the same CodeCommit repo, but this time select the new buildspec file. Use a Python 3.3.6 builder – either the AWS or LI Instrumented version.
  5. Click Continue
  6. Click Save
  7. Run the build, again on the master branch.
  8. If everything runs successfully, a new image should appear in the ECR registry with a -layered suffix. This is the instrumented image.

Run instrumented container image

When the instrumented container is now run — in ECS, Fargate, or elsewhere — it will log data back to the Layered Insight runtime console. It’s appearance in the console can be modified by setting the LI_APPLICATIONNAME and LI_LOCATION environment variables when running the container.

Conclusion

In the above blog we have provided you steps needed to embed governance and runtime security in your build pipelines running on AWS CodeBuild using Layered Insight.

 

 

 

Get Started with Blockchain Using the new AWS Blockchain Templates

Post Syndicated from Jeff Barr original https://aws.amazon.com/blogs/aws/get-started-with-blockchain-using-the-new-aws-blockchain-templates/

Many of today’s discussions around blockchain technology remind me of the classic Shimmer Floor Wax skit. According to Dan Aykroyd, Shimmer is a dessert topping. Gilda Radner claims that it is a floor wax, and Chevy Chase settles the debate and reveals that it actually is both! Some of the people that I talk to see blockchains as the foundation of a new monetary system and a way to facilitate international payments. Others see blockchains as a distributed ledger and immutable data source that can be applied to logistics, supply chain, land registration, crowdfunding, and other use cases. Either way, it is clear that there are a lot of intriguing possibilities and we are working to help our customers use this technology more effectively.

We are launching AWS Blockchain Templates today. These templates will let you launch an Ethereum (either public or private) or Hyperledger Fabric (private) network in a matter of minutes and with just a few clicks. The templates create and configure all of the AWS resources needed to get you going in a robust and scalable fashion.

Launching a Private Ethereum Network
The Ethereum template offers two launch options. The ecs option creates an Amazon ECS cluster within a Virtual Private Cloud (VPC) and launches a set of Docker images in the cluster. The docker-local option also runs within a VPC, and launches the Docker images on EC2 instances. The template supports Ethereum mining, the EthStats and EthExplorer status pages, and a set of nodes that implement and respond to the Ethereum RPC protocol. Both options create and make use of a DynamoDB table for service discovery, along with Application Load Balancers for the status pages.

Here are the AWS Blockchain Templates for Ethereum:

I start by opening the CloudFormation Console in the desired region and clicking Create Stack:

I select Specify an Amazon S3 template URL, enter the URL of the template for the region, and click Next:

I give my stack a name:

Next, I enter the first set of parameters, including the network ID for the genesis block. I’ll stick with the default values for now:

I will also use the default values for the remaining network parameters:

Moving right along, I choose the container orchestration platform (ecs or docker-local, as I explained earlier) and the EC2 instance type for the container nodes:

Next, I choose my VPC and the subnets for the Ethereum network and the Application Load Balancer:

I configure my keypair, EC2 security group, IAM role, and instance profile ARN (full information on the required permissions can be found in the documentation):

The Instance Profile ARN can be found on the summary page for the role:

I confirm that I want to deploy EthStats and EthExplorer, choose the tag and version for the nested CloudFormation templates that are used by this one, and click Next to proceed:

On the next page I specify a tag for the resources that the stack will create, leave the other options as-is, and click Next:

I review all of the parameters and options, acknowledge that the stack might create IAM resources, and click Create to build my network:

The template makes use of three nested templates:

After all of the stacks have been created (mine took about 5 minutes), I can select JeffNet and click the Outputs tab to discover the links to EthStats and EthExplorer:

Here’s my EthStats:

And my EthExplorer:

If I am writing apps that make use of my private network to store and process smart contracts, I would use the EthJsonRpcUrl.

Stay Tuned
My colleagues are eager to get your feedback on these new templates and plan to add new versions of the frameworks as they become available.

— Jeff;

 

Amazon ECS Service Discovery

Post Syndicated from Randall Hunt original https://aws.amazon.com/blogs/aws/amazon-ecs-service-discovery/

Amazon ECS now includes integrated service discovery. This makes it possible for an ECS service to automatically register itself with a predictable and friendly DNS name in Amazon Route 53. As your services scale up or down in response to load or container health, the Route 53 hosted zone is kept up to date, allowing other services to lookup where they need to make connections based on the state of each service. You can see a demo of service discovery in an imaginary social networking app over at: https://servicediscovery.ranman.com/.

Service Discovery


Part of the transition to microservices and modern architectures involves having dynamic, autoscaling, and robust services that can respond quickly to failures and changing loads. Your services probably have complex dependency graphs of services they rely on and services they provide. A modern architectural best practice is to loosely couple these services by allowing them to specify their own dependencies, but this can be complicated in dynamic environments as your individual services are forced to find their own connection points.

Traditional approaches to service discovery like consul, etcd, or zookeeper all solve this problem well, but they require provisioning and maintaining additional infrastructure or installation of agents in your containers or on your instances. Previously, to ensure that services were able to discover and connect with each other, you had to configure and run your own service discovery system or connect every service to a load balancer. Now, you can enable service discovery for your containerized services in the ECS console, AWS CLI, or using the ECS API.

Introducing Amazon Route 53 Service Registry and Auto Naming APIs

Amazon ECS Service Discovery works by communicating with the Amazon Route 53 Service Registry and Auto Naming APIs. Since we haven’t talked about it before on this blog, I want to briefly outline how these Route 53 APIs work. First, some vocabulary:

  • Namespaces – A namespace specifies a domain name you want to route traffic to (e.g. internal, local, corp). You can think of it as a logical boundary between which services should be able to discover each other. You can create a namespace with a call to the aws servicediscovery create-private-dns-namespace command or in the ECS console. Namespaces are roughly equivalent to hosted zones in Route 53. A namespace contains services, our next vocabulary word.
  • Service – A service is a specific application or set of applications in your namespace like “auth”, “timeline”, or “worker”. A service contains service instances.
  • Service Instance – A service instance contains information about how Route 53 should respond to DNS queries for a resource.

Route 53 provides APIs to create: namespaces, A records per task IP, and SRV records per task IP + port.

When we ask Route 53 for something like: worker.corp we should get back a set of possible IPs that could fulfill that request. If the application we’re connecting to exposes dynamic ports then the calling application can easily query the SRV record to get more information.

ECS service discovery is built on top of the Route 53 APIs and manages all of the underlying API calls for you. Now that we understand how the service registry, works lets take a look at the ECS side to see service discovery in action.

Amazon ECS Service Discovery

Let’s launch an application with service discovery! First, I’ll create two task definitions: “flask-backend” and “flask-worker”. Both are simple AWS Fargate tasks with a single container serving HTTP requests. I’ll have flask-backend ask worker.corp to do some work and I’ll return the response as well as the address Route 53 returned for worker. Something like the code below:

@app.route("/")
namespace = os.getenv("namespace")
worker_host = "worker" + namespace
def backend():
    r = requests.get("http://"+worker_host)
    worker = socket.gethostbyname(worker_host)
    return "Worker Message: {]\nFrom: {}".format(r.content, worker)

 

Now, with my containers and task definitions in place, I’ll create a service in the console.

As I move to step two in the service wizard I’ll fill out the service discovery section and have ECS create a new namespace for me.

I’ll also tell ECS to monitor the health of the tasks in my service and add or remove them from Route 53 as needed. Then I’ll set a TTL of 10 seconds on the A records we’ll use.

I’ll repeat those same steps for my “worker” service and after a minute or so most of my tasks should be up and running.

Over in the Route 53 console I can see all the records for my tasks!

We can use the Route 53 service discovery APIs to list all of our available services and tasks and programmatically reach out to each one. We could easily extend to any number of services past just backend and worker. I’ve created a simple demo of an imaginary social network with services like “auth”, “feed”, “timeline”, “worker”, “user” and more here: https://servicediscovery.ranman.com/. You can see the code used to run that page on github.

Available Now
Amazon ECS service discovery is available now in US East (N. Virginia), US East (Ohio), US West (Oregon), and EU (Ireland). AWS Fargate is currently only available in US East (N. Virginia). When you use ECS service discovery, you pay for the Route 53 resources that you consume, including each namespace that you create, and for the lookup queries your services make. Container level health checks are provided at no cost. For more information on pricing check out the documentation.

Please let us know what you’ll be building or refactoring with service discovery either in the comments or on Twitter!

– Randall

 

P.S. Every blog post I write is made with a tremendous amount of help from numerous AWS colleagues. To everyone that helped build service discovery across all of our teams – thank you :)!

Power from wind: Open data on AWS

Post Syndicated from Caleb Phillips original https://aws.amazon.com/blogs/big-data/power-from-wind-open-data-on-aws/

Data that describe processes in a spatial context are everywhere in our day-to-day lives and they dominate big data problems. Map data, for instance, whether describing networks of roads or remote sensing data from satellites, get us where we need to go. Atmospheric data from simulations and sensors underlie our weather forecasts and climate models. Devices and sensors with GPS can provide a spatial context to nearly all mobile data.

In this post, we introduce the WIND toolkit, a huge (500 TB), open weather model dataset that’s available to the world on Amazon’s cloud services. We walk through how to access this data and some of the open-source software developed to make it easily accessible. Our solution considers a subset of geospatial data that exist on a grid (raster) and explores ways to provide access to large-scale raster data from weather models. The solution uses foundational AWS services and the Hierarchical Data Format (HDF), a well adopted format for scientific data.

The approach developed here can be extended to any data that fit in an HDF5 file, which can describe sparse and dense vectors and matrices of arbitrary dimensions. This format is already popular within the physical sciences for both experimental and simulation data. We discuss solutions to gridded data storage for a massive dataset of public weather model outputs called the Wind Integration National Dataset (WIND) toolkit. We also highlight strategies that are general to other large geospatial data management problems.

Wind Integration National Dataset

As variable renewable power penetration levels increase in power systems worldwide, the importance of renewable integration studies to ensure continued economic and reliable operation of the power grid is also increasing. The WIND toolkit is the largest freely available grid integration dataset to date.

The WIND toolkit was developed by 3TIER by Vaisala. They were under a subcontract to the National Renewable Energy Laboratory (NREL) to support studies on integration of wind energy into the existing US grid. NREL is a part of a network of national laboratories for the US Department of Energy and has a mission to advance the science and engineering of energy efficiency, sustainable transportation, and renewable power technologies.

The toolkit has been used by consultants, research groups, and universities worldwide to support grid integration studies. Less traditional uses also include resource assessments for wind plants (such as those powering Amazon data centers), and studying the effects of weather on California condor migrations in the Baja peninsula.

The diversity of applications highlights the value of accessible, open public data. Yet, there’s a catch: the dataset is huge. The WIND toolkit provides simulated atmospheric (weather) data at a two-km spatial resolution and five-minute temporal resolution at multiple heights for seven years. The entire dataset is half a petabyte (500 TB) in size and is stored in the NREL High Performance Computing data center in Golden, Colorado. Making this dataset publicly available easily and in a cost-effective manner is a major challenge.

As other laboratories and public institutions work to release their data to the world, they may face similar challenges to those that we experienced. Some prior, well-intentioned efforts to release huge datasets as-is have resulted in data resources that are technically available but fundamentally unusable. They may be stored in an unintuitive format or indexed and organized to support only a subset of potential uses. Downloading hundreds of terabytes of data is often impractical. Most users don’t have access to a big data cluster (or super computer) to slice and dice the data as they need after it’s downloaded.

We aim to provide a large amount of data (50 terabytes) to the public in a way that is efficient, scalable, and easy to use. In many cases, researchers can access these huge cloud-located datasets using the same software and algorithms they have developed for smaller datasets stored locally. Only the pieces of data they need for their individual analysis must be downloaded. To make this work in practice, we worked with the HDF Group and have built upon their forthcoming Highly Scalable Data Service.

In the rest of this post, we discuss how the HSDS software was developed to use Amazon EC2 and Amazon S3 resources to provide convenient and scalable access to these huge geospatial datasets. We describe how the HSDS service has been put to work for the WIND Toolkit dataset and demonstrate how to access it using the h5pyd Python library and the REST API. We conclude with information about our ongoing work to release more ‘open’ datasets to the public using AWS services, and ways to improve and extend the HSDS with newer Amazon services like Amazon ECS and AWS Lambda.

Developing a scalable service for big geospatial data

The HDF5 file format and API have been used for many years and is an effective means of storing large scientific datasets.  For example, NASA’s Earth Observing System (EOS) satellites collect more than 16 TBs of data per day using HDF5.

With the rise of the cloud, there are new challenges and opportunities to rethink how HDF5 can be enhanced to work effectively as a component in a cloud-native architecture.  For the HDF Group, working with NREL has been a great opportunity to put ideas into practice with a production-size dataset.

An HDF5 file consists of a directed graph of group and dataset objects.  Datasets can be thought of as a multidimensional array with support for user-defined metadata tags and compression.  Typical operations on datasets would be reading or writing data to a regular subregion (a hyperslab) or reading and writing individual elements (a point selection).   Also, group and dataset objects may each contain an arbitrary number of the user-defined metadata elements known as attributes.

Many people have used the HDF library in applications developed or ported to run on EC2 instances, but there are a number of constraints that often prove problematic:

  • The HDF5 library can’t read directly from HDF5 files stored as S3 objects. The entire file (often many GB in size) would need to be copied to local storage before the first byte can be read. Also, the instance must be configured with the appropriately sized EBS volume)
  • The HDF library only has access to the computational resources of the instance itself (as opposed to a cluster of instances), so many operations are bottlenecked by the library.
  • Any modifications to the HDF5 file would somehow have to be synchronized with changes that other instances have made to same file before writing back to S3.

Using a pattern common to many offerings from AWS, the solution to these constraints is to develop a service framework around the HDF data model. Using this model, the HDF Group has created the Highly Scalable Data Service (HSDS) that provides all the functionality that traditionally was provided by the HDF5 library.  By using the service, you don’t need to manage your own file volumes, but can just read and write whatever data that you need.

Because the service manages the actual data persistence to a durable medium (S3, in this case), you don’t need to worry about disk management. Simply stream the data you need from the service as you need it.  Secondly, putting the functionality behind a service allows some tricks to increase performance (described in more detail later). And lastly, HSDS allows any number of clients to access the data at the same time, enabling HDF5 to be used as a coordination mechanism for multiple readers and writers.

In designing the HSDS architecture, we gave much thought to how to achieve scalability of the HSDS service.  For accessing HDF5 data, there are two different types of scaling to consider:

  • Multiple clients making many requests to the service
  • Single requests that require a significant amount of data processing

To deal with the first scaling challenge, as with most services, we considered how the service responds as the request rate increases.  AWS provides some great tools that help in this regard:

  • Auto Scaling groups
  • Elastic Load Balancing load balancers
  • The ability of S3 to handle large aggregate throughput rates

By using a cluster of EC2 instances behind a load balancer, you can handle different client loads in a cost-effective manner.

The second scaling challenge concerns single requests that would take significant processing time with just one compute node.  One example of this from the WIND toolkit would be extracting all the values in the seven-year time span for a given geographic point and dataset.

In HDF5, large datasets are typically stored as “chunks”; that is, a regular partition of the array.  In HSDS, each chunk is stored as a binary object in S3.  The sequential approach to retrieving the time series values would be for the service to read each chunk needed from S3, extract the needed elements, and go on to the next chunk.  In this case, that would involve processing 2557 chunks, and would be quite slow.

Fortunately, with HSDS, you can speed this up quite a bit by exploiting the compute and I/O capabilities of the cluster.  Upon receiving the request, the receiving node can use other nodes in the cluster to read different portions of the selection.  With multiple nodes reading from S3 in parallel, performance improves as the cluster size increases.

The diagram below illustrates how this works in simplified case of four chunks and four nodes.

This architecture has worked in well in practice.  In testing with the WIND toolkit and time series extraction, we observed a request latency of ~60 seconds using four nodes vs. ~5 seconds with 40 nodes. Performance roughly scales with the size of the cluster.

A planned enhancement to this is to use AWS Lambda for the worker processing.  This enables 1000-way parallel reads at a reasonable cost, as you only pay for the milliseconds of CPU time used with AWS Lambda.

Public access to atmospheric data using HSDS and AWS

An early challenge in releasing the WIND toolkit data was in deciding how to subset the data for different use cases. In general, few researchers need access to the entire 0.5 PB of data and a great deal of efficiency and cost reduction can be gained by making directed constituent datasets.

NREL grid integration researchers initially extracted a 2-TB subset by selecting 120,000 points where the wind resource seemed appropriate for development. They also chose only those data important for wind applications (100-m wind speed, converted to power), the most interesting locations for those performing grid studies. To support the remaining users who needed more data resolution, we down-sampled the data to a 60-minute temporal resolution, keeping all the other variables and spatial resolution intact. This reduced dataset is 50 TB of data describing 30+ atmospheric variables of data for 7 years at a 60-minute temporal resolution.

The WindViz browser-based Gridded Wind Toolkit Visualizer was created as an example implementation of the HSDS REST API in JavaScript. The visualizer is written in the style of ECMAScript 2016 using a modern development toolchain that includes webpack and Babel. The source code is available through our GitHub repository. The demo page is hosted via GitHub pages, and we use a cross-origin AJAX request to fetch data from the HSDS service running on the EC2 infrastructure. The visualizer can be used to explore the gridded wind toolkit data on a map. Achieve full spatial resolution by zooming in to a specific region.

Programmatic access is possible using the h5pyd Python library, a distributed analog to the widely used h5py library. Users interact with the datasets (variables) and slice the data from its (time x longitude x latitude) cube form as they see fit.

Examples and use cases are described in a set of Jupyter notebooks and available on GitHub:

NREL/hsds-examples

To run these notebooks on an EC2 instance in the Oregon Region, run the following commands:

$ sudo yum install git gcc
$ sudo pip install –user jupyter
$ pip install --user git+http://github.com/HDFGroup/h5pyd.git
$ git clone https://github.com/NREL/hsds-examples.git
$ cd hsds-examples
$ jupyter notebook

Now you have a Jupyter notebook server running on your EC2 server.

From your laptop, create an SSH tunnel:

$ ssh –L 8888:localhost:8888 (IP address of the EC2 server)

Now, you can browse to localhost:8888 using the correct token, and interact with the notebooks as if they were local. Within the directory, there are examples for accessing the HSDS API and plotting wind and weather data using matplotlib.

Controlling access and defraying costs

A final concern is rate limiting and access control. Although the HSDS service is scalable and relatively robust, we had a few practical concerns:

  • How can we protect from malicious or accidental use that may lead to high egress fees (for example, someone who attempts to repeatedly download the entire dataset from S3)?
  • How can we keep track of who is using the data both to document the value of the data resource and to justify the costs?
  • If costs become too high, can we charge for some or all API use to help cover the costs?

To approach these problems, we investigated using Amazon API Gateway and its simplified integration with the AWS Marketplace for SaaS monetization as well as third-party API proxies.

In the end, we chose to use API Umbrella due to its close involvement with http://data.gov. While AWS Marketplace is a compelling option for future datasets, the decision was made to keep this dataset entirely open, at least for now. As community use and associated costs grow, we’ll likely revisit Marketplace. Meanwhile, API Umbrella provides controls for rate limiting and API key registration out of the box and was simple to implement as a front-end proxy to HSDS. Those applications that may want to charge for API use can accomplish a similar strategy using Amazon API Gateway and AWS Marketplace.

Ongoing work and other resources

As NREL and other government research labs, municipalities, and organizations try to share data with the public, we expect many of you will face similar challenges to those we have tried to approach with the architecture described in this post. Providing large datasets is one challenge. Doing so in a way that is affordable and convenient for users is an entirely more difficult goal. Using AWS cloud-native services and the existing foundation of the HDF file format has allowed us to tackle that challenge in a meaningful way.

 

Additional Reading

If you found this post useful, be sure to check out Perform Near Real-time Analytics on Streaming Data with Amazon Kinesis and Amazon Elasticsearch Service, Analyze OpenFDA Data in R with Amazon S3 and Amazon Athena and Querying OpenStreetMap with Amazon Athena.

 

About the Authors

Dr. Caleb Phillips is a senior scientist with the Data Analysis and Visualization Group within the Computational Sciences Center at the National Renewable Energy Laboratory. Caleb comes from a background in computer science systems, applied statistics, computational modeling, and optimization. His work at NREL spans the breadth of renewable energy technologies and focuses on applying modern data science techniques to data problems at scale.

 

 

 

Dr. Caroline Draxl is a senior scientist at NREL. She supports the research and modeling activities of the US Department of Energy from mesoscale to wind plant scale. Caroline uses mesoscale models to research wind resources in various countries, and participates in on- and offshore boundary layer research and in the coupling of the mesoscale flow features (kilometer scale) to the microscale (tens of meters). She holds a M.S. degree in Meteorology and Geophysics from the University of Innsbruck, Austria, and a PhD in Meteorology from the Technical University of Denmark.

 

 

 

John Readey has been a Senior Architect at The HDF Group since he joined in June 2014.  His interests include web services related to HDF, applications that support the use of HDF and data visualization.Before joining The HDF Group, John worked at Amazon.com from 2006–2014 where he developed service-based systems for eCommerce and AWS.

 

 

 

Jordan Perr-Sauer is an RPP intern with the Data Analysis and Visualization Group within the Computational Sciences Center at the National Renewable Energy Laboratory. Jordan hopes to use his professional background in software engineering and his academic training in applied mathematics to solve the challenging problems facing America and the world.

Central Logging in Multi-Account Environments

Post Syndicated from matouk original https://aws.amazon.com/blogs/architecture/central-logging-in-multi-account-environments/

Centralized logging is often required in large enterprise environments for a number of reasons, ranging from compliance and security to analytics and application-specific needs.

I’ve seen that in a multi-account environment, whether the accounts belong to the same line of business or multiple business units, collecting logs in a central, dedicated logging account is an established best practice. It helps security teams detect malicious activities both in real-time and during incident response. It provides protection to log data in case it is accidentally or intentionally deleted. It also helps application teams correlate and analyze log data across multiple application tiers.

This blog post provides a solution and building blocks to stream Amazon CloudWatch log data across accounts. In a multi-account environment this repeatable solution could be deployed multiple times to stream all relevant Amazon CloudWatch log data from all accounts to a centralized logging account.

Solution Summary 

The solution uses Amazon Kinesis Data Streams and a log destination to set up an endpoint in the logging account to receive streamed logs and uses Amazon Kinesis Data Firehose to deliver log data to the Amazon Simple Storage Solution (S3) bucket. Application accounts will subscribe to stream all (or part) of their Amazon CloudWatch logs to a defined destination in the logging account via subscription filters.

Below is a diagram illustrating how the various services work together.


In logging an account, a Kinesis Data Stream is created to receive streamed log data and a log destination is created to facilitate remote streaming, configured to use the Kinesis Data Stream as its target.

The Amazon Kinesis Data Firehose stream is created to deliver log data from the data stream to S3. The delivery stream uses a generic AWS Lambda function for data validation and transformation.

In each application account, a subscription filter is created between each Amazon CloudWatch log group and the destination created for this log group in the logging account.

The following steps are involved in setting up the central-logging solution:

  1. Create an Amazon S3 bucket for your central logging in the logging account
  2. Create an AWS Lambda function for log data transformation and decoding in logging account
  3. Create a central logging stack as a logging-account destination ready to receive streamed logs and deliver them to S3
  4. Create a subscription in application accounts to deliver logs from a specific CloudWatch log group to the logging account destination
  5. Create Amazon Athena tables to query and analyze log data in your logging account

Creating a log destination in your logging account

In this section, we will setup the logging account side of the solution, providing detail on the list above. The example I use is for the us-east-1 region, however any region where required services are available could be used.

It’s important to note that your logging-account destination and application-account subscription must be in the same region. You can deploy the solution multiple times to create destinations in all required regions if application accounts use multiple regions.

Step 1: Create an S3 bucket

Use the CloudFormation template below to create S3 bucket in logging account. This template also configures the bucket to archive log data to Glacier after 60 days.


{
  "AWSTemplateFormatVersion":"2010-09-09",
  "Description": "CF Template to create S3 bucket for central logging",
  "Parameters":{

    "BucketName":{
      "Type":"String",
      "Default":"",
      "Description":"Central logging bucket name"
    }
  },
  "Resources":{
                        
   "CentralLoggingBucket" : {
      "Type" : "AWS::S3::Bucket",
      "Properties" : {
        "BucketName" : {"Ref": "BucketName"},
        "LifecycleConfiguration": {
            "Rules": [
                {
                  "Id": "ArchiveToGlacier",
                  "Prefix": "",
                  "Status": "Enabled",
                  "Transitions":[{
                      "TransitionInDays": "60",
                      "StorageClass": "GLACIER"
                  }]
                }
            ]
        }
      }
    }

  },
  "Outputs":{
    "CentralLogBucket":{
    	"Description" : "Central log bucket",
    	"Value" : {"Ref": "BucketName"} ,
    	"Export" : { "Name" : "CentralLogBucketName"}
    }
  }
} 

To create your central-logging bucket do the following:

  1. Save the template file to your local developer machine as “central-log-bucket.json”
  2. From the CloudFormation console, select “create new stack” and import the file “central-log-bucket.json”
  3. Fill in the parameters and complete stack creation steps (as indicated in the screenshot below)
  4. Verify the bucket has been created successfully and take a note of the bucket name

Step 2: Create data processing Lambda function

Use the template below to create a Lambda function in your logging account that will be used by Amazon Firehose for data transformation during the delivery process to S3. This function is based on the AWS Lambda kinesis-firehose-cloudwatch-logs-processor blueprint.

The function could be created manually from the blueprint or using the cloudformation template below. To find the blueprint navigate to Lambda -> Create -> Function -> Blueprints

This function will unzip the event message, parse it and verify that it is a valid CloudWatch log event. Additional processing can be added if needed. As this function is generic, it could be reused by all log-delivery streams.

{
  "AWSTemplateFormatVersion":"2010-09-09",
  "Description": "Create cloudwatch data processing lambda function",
  "Resources":{
      
    "LambdaRole": {
        "Type": "AWS::IAM::Role",
        "Properties": {
            "AssumeRolePolicyDocument": {
                "Version": "2012-10-17",
                "Statement": [
                    {
                        "Effect": "Allow",
                        "Principal": {
                            "Service": "lambda.amazonaws.com"
                        },
                        "Action": "sts:AssumeRole"
                    }
                ]
            },
            "Path": "/",
            "Policies": [
                {
                    "PolicyName": "firehoseCloudWatchDataProcessing",
                    "PolicyDocument": {
                        "Version": "2012-10-17",
                        "Statement": [
                            {
                                "Effect": "Allow",
                                "Action": [
                                    "logs:CreateLogGroup",
                                    "logs:CreateLogStream",
                                    "logs:PutLogEvents"
                                ],
                                "Resource": "arn:aws:logs:*:*:*"
                            }
                        ]
                    }
                }
            ]
        }
    },
      
    "FirehoseDataProcessingFunction": {
        "Type": "AWS::Lambda::Function",
        "Properties": {
            "Handler": "index.handler",
            "Role": {"Fn::GetAtt": ["LambdaRole","Arn"]},
            "Description": "Firehose cloudwatch data processing",
            "Code": {
                "ZipFile" : { "Fn::Join" : ["\n", [
                  "'use strict';",
                  "const zlib = require('zlib');",
                  "function transformLogEvent(logEvent) {",
                  "       return Promise.resolve(`${logEvent.message}\n`);",
                  "}",
                  "exports.handler = (event, context, callback) => {",
                  "    Promise.all(event.records.map(r => {",
                  "        const buffer = new Buffer(r.data, 'base64');",
                  "        const decompressed = zlib.gunzipSync(buffer);",
                  "        const data = JSON.parse(decompressed);",
                  "        if (data.messageType !== 'DATA_MESSAGE') {",
                  "            return Promise.resolve({",
                  "                recordId: r.recordId,",
                  "                result: 'ProcessingFailed',",
                  "            });",
                  "         } else {",
                  "            const promises = data.logEvents.map(transformLogEvent);",
                  "            return Promise.all(promises).then(transformed => {",
                  "                const payload = transformed.reduce((a, v) => a + v, '');",
                  "                const encoded = new Buffer(payload).toString('base64');",
                  "                console.log('---------------payloadv2:'+JSON.stringify(payload, null, 2));",
                  "                return {",
                  "                    recordId: r.recordId,",
                  "                    result: 'Ok',",
                  "                    data: encoded,",
                  "                };",
                  "           });",
                  "        }",
                  "    })).then(recs => callback(null, { records: recs }));",
                    "};"

                ]]}
            },
            "Runtime": "nodejs6.10",
            "Timeout": "60"
        }
    }

  },
  "Outputs":{
   "Function" : {
      "Description": "Function ARN",
      "Value": {"Fn::GetAtt": ["FirehoseDataProcessingFunction","Arn"]},
      "Export" : { "Name" : {"Fn::Sub": "${AWS::StackName}-Function" }}
    }
  }
}

To create the function follow the steps below:

  1. Save the template file as “central-logging-lambda.json”
  2. Login to logging account and, from the CloudFormation console, select “create new stack”
  3. Import the file “central-logging-lambda.json” and click next
  4. Follow the steps to create the stack and verify successful creation
  5. Take a note of Lambda function arn from the output section

Step 3: Create log destination in logging account

Log destination is used as the target of a subscription from application accounts, log destination can be shared between multiple subscriptions however according to the architecture suggested in this solution all logs streamed to the same destination will be stored in the same S3 location, if you would like to store log data in different hierarchy or in a completely different bucket you need to create separate destinations.

As noted previously, your destination and subscription have to be in the same region

Use the template below to create destination stack in logging account.

{
  "AWSTemplateFormatVersion":"2010-09-09",
  "Description": "Create log destination and required resources",
  "Parameters":{

    "LogBucketName":{
      "Type":"String",
      "Default":"central-log-do-not-delete",
      "Description":"Destination logging bucket"
    },
    "LogS3Location":{
      "Type":"String",
      "Default":"<BU>/<ENV>/<SOURCE_ACCOUNT>/<LOG_TYPE>/",
      "Description":"S3 location for the logs streamed to this destination; example marketing/prod/999999999999/flow-logs/"
    },
    "ProcessingLambdaARN":{
      "Type":"String",
      "Default":"",
      "Description":"CloudWatch logs data processing function"
    },
    "SourceAccount":{
      "Type":"String",
      "Default":"",
      "Description":"Source application account number"
    }
  },
    
  "Resources":{
    "MyStream": {
      "Type": "AWS::Kinesis::Stream",
      "Properties": {
        "Name": {"Fn::Join" : [ "", [{ "Ref" : "AWS::StackName" },"-Stream"] ]},
        "RetentionPeriodHours" : 48,
        "ShardCount": 1,
        "Tags": [
          {
            "Key": "Solution",
            "Value": "CentralLogging"
          }
       ]
      }
    },
    "LogRole" : {
      "Type"  : "AWS::IAM::Role",
      "Properties" : {
          "AssumeRolePolicyDocument" : {
              "Statement" : [ {
                  "Effect" : "Allow",
                  "Principal" : {
                      "Service" : [ {"Fn::Join": [ "", [ "logs.", { "Ref": "AWS::Region" }, ".amazonaws.com" ] ]} ]
                  },
                  "Action" : [ "sts:AssumeRole" ]
              } ]
          },         
          "Path" : "/service-role/"
      }
    },
      
    "LogRolePolicy" : {
        "Type" : "AWS::IAM::Policy",
        "Properties" : {
            "PolicyName" : {"Fn::Join" : [ "", [{ "Ref" : "AWS::StackName" },"-LogPolicy"] ]},
            "PolicyDocument" : {
              "Version": "2012-10-17",
              "Statement": [
                {
                  "Effect": "Allow",
                  "Action": ["kinesis:PutRecord"],
                  "Resource": [{ "Fn::GetAtt" : ["MyStream", "Arn"] }]
                },
                {
                  "Effect": "Allow",
                  "Action": ["iam:PassRole"],
                  "Resource": [{ "Fn::GetAtt" : ["LogRole", "Arn"] }]
                }
              ]
            },
            "Roles" : [ { "Ref" : "LogRole" } ]
        }
    },
      
    "LogDestination" : {
      "Type" : "AWS::Logs::Destination",
      "DependsOn" : ["MyStream","LogRole","LogRolePolicy"],
      "Properties" : {
        "DestinationName": {"Fn::Join" : [ "", [{ "Ref" : "AWS::StackName" },"-Destination"] ]},
        "RoleArn": { "Fn::GetAtt" : ["LogRole", "Arn"] },
        "TargetArn": { "Fn::GetAtt" : ["MyStream", "Arn"] },
        "DestinationPolicy": { "Fn::Join" : ["",[
		
				"{\"Version\" : \"2012-10-17\",\"Statement\" : [{\"Effect\" : \"Allow\",",
                " \"Principal\" : {\"AWS\" : \"", {"Ref":"SourceAccount"} ,"\"},",
                "\"Action\" : \"logs:PutSubscriptionFilter\",",
                " \"Resource\" : \"", 
                {"Fn::Join": [ "", [ "arn:aws:logs:", { "Ref": "AWS::Region" }, ":" ,{ "Ref": "AWS::AccountId" }, ":destination:",{ "Ref" : "AWS::StackName" },"-Destination" ] ]}  ,"\"}]}"

			]]}
          
          
      }
    },
      
    "S3deliveryStream": {
      "DependsOn": ["S3deliveryRole", "S3deliveryPolicy"],
      "Type": "AWS::KinesisFirehose::DeliveryStream",
      "Properties": {
        "DeliveryStreamName": {"Fn::Join" : [ "", [{ "Ref" : "AWS::StackName" },"-DeliveryStream"] ]},
        "DeliveryStreamType": "KinesisStreamAsSource",
        "KinesisStreamSourceConfiguration": {
            "KinesisStreamARN": { "Fn::GetAtt" : ["MyStream", "Arn"] },
            "RoleARN": {"Fn::GetAtt" : ["S3deliveryRole", "Arn"] }
        },
        "ExtendedS3DestinationConfiguration": {
          "BucketARN": {"Fn::Join" : [ "", ["arn:aws:s3:::",{"Ref":"LogBucketName"}] ]},
          "BufferingHints": {
            "IntervalInSeconds": "60",
            "SizeInMBs": "50"
          },
          "CompressionFormat": "UNCOMPRESSED",
          "Prefix": {"Ref": "LogS3Location"},
          "RoleARN": {"Fn::GetAtt" : ["S3deliveryRole", "Arn"] },
          "ProcessingConfiguration" : {
              "Enabled": "true",
              "Processors": [
              {
                "Parameters": [ 
                { 
                    "ParameterName": "LambdaArn",
                    "ParameterValue": {"Ref":"ProcessingLambdaARN"}
                }],
                "Type": "Lambda"
              }]
          }
        }

      }
    },
      
    "S3deliveryRole": {
      "Type": "AWS::IAM::Role",
      "Properties": {
        "AssumeRolePolicyDocument": {
          "Version": "2012-10-17",
          "Statement": [
            {
              "Sid": "",
              "Effect": "Allow",
              "Principal": {
                "Service": "firehose.amazonaws.com"
              },
              "Action": "sts:AssumeRole",
              "Condition": {
                "StringEquals": {
                  "sts:ExternalId": {"Ref":"AWS::AccountId"}
                }
              }
            }
          ]
        }
      }
    },
      
    "S3deliveryPolicy": {
      "Type": "AWS::IAM::Policy",
      "Properties": {
        "PolicyName": {"Fn::Join" : [ "", [{ "Ref" : "AWS::StackName" },"-FirehosePolicy"] ]},
        "PolicyDocument": {
          "Version": "2012-10-17",
          "Statement": [
            {
              "Effect": "Allow",
              "Action": [
                "s3:AbortMultipartUpload",
                "s3:GetBucketLocation",
                "s3:GetObject",
                "s3:ListBucket",
                "s3:ListBucketMultipartUploads",
                "s3:PutObject"
              ],
              "Resource": [
                {"Fn::Join": ["", [ {"Fn::Join" : [ "", ["arn:aws:s3:::",{"Ref":"LogBucketName"}] ]}]]},
                {"Fn::Join": ["", [ {"Fn::Join" : [ "", ["arn:aws:s3:::",{"Ref":"LogBucketName"}] ]}, "*"]]}
              ]
            },
            {
              "Effect": "Allow",
              "Action": [
                "lambda:InvokeFunction",
                "lambda:GetFunctionConfiguration",
                "logs:PutLogEvents",
                "kinesis:DescribeStream",
                "kinesis:GetShardIterator",
                "kinesis:GetRecords",
                "kms:Decrypt"
              ],
              "Resource": "*"
            }
          ]
        },
        "Roles": [{"Ref": "S3deliveryRole"}]
      }
    }

  },
  "Outputs":{
      
   "Destination" : {
      "Description": "Destination",
      "Value": {"Fn::Join": [ "", [ "arn:aws:logs:", { "Ref": "AWS::Region" }, ":" ,{ "Ref": "AWS::AccountId" }, ":destination:",{ "Ref" : "AWS::StackName" },"-Destination" ] ]},
      "Export" : { "Name" : {"Fn::Sub": "${AWS::StackName}-Destination" }}
    }

  }
}

To create log your destination and all required resources, follow these steps:

  1. Save your template as “central-logging-destination.json”
  2. Login to your logging account and, from the CloudFormation console, select “create new stack”
  3. Import the file “central-logging-destination.json” and click next
  4. Fill in the parameters to configure the log destination and click Next
  5. Follow the default steps to create the stack and verify successful creation
    1. Bucket name is the same as in the “create central logging bucket” step
    2. LogS3Location is the directory hierarchy for saving log data that will be delivered to this destination
    3. ProcessingLambdaARN is as created in “create data processing Lambda function” step
    4. SourceAccount is the application account number where the subscription will be created
  6. Take a note of destination ARN as it appears in outputs section as you did above.

Step 4: Create the log subscription in your application account

In this section, we will create the subscription filter in one of the application accounts to stream logs from the CloudWatch log group to the log destination that was created in your logging account.

Create log subscription filter

The subscription filter is created between the CloudWatch log group and a destination endpoint. Asubscription could be filtered to send part (or all) of the logs in the log group. For example,you can create a subscription filter to stream only flow logs with status REJECT.

Use the CloudFormation template below to create subscription filter. Subscription filter and log destination must be in the same region.

{
  "AWSTemplateFormatVersion":"2010-09-09",
  "Description": "Create log subscription filter for a specific Log Group",
  "Parameters":{

    "DestinationARN":{
      "Type":"String",
      "Default":"",
      "Description":"ARN of logs destination"
    },
    "LogGroupName":{
      "Type":"String",
      "Default":"",
      "Description":"Name of LogGroup to forward logs from"
    },
    "FilterPattern":{
      "Type":"String",
      "Default":"",
      "Description":"Filter pattern to filter events to be sent to log destination; Leave empty to send all logs"
    }
  },
    
  "Resources":{
    "SubscriptionFilter" : {
      "Type" : "AWS::Logs::SubscriptionFilter",
      "Properties" : {
        "LogGroupName" : { "Ref" : "LogGroupName" },
        "FilterPattern" : { "Ref" : "FilterPattern" },
        "DestinationArn" : { "Ref" : "DestinationARN" }
      }
    }
  }
}

To create a subscription filter for one of CloudWatch log groups in your application account, follow the steps below:

  1. Save the template as “central-logging-subscription.json”
  2. Login to your application account and, from the CloudFormation console, select “create new stack”
  3. Select the file “central-logging-subscription.json” and click next
  4. Fill in the parameters as appropriate to your environment as you did above
    a.  DestinationARN is the value of obtained in “create log destination in logging account” step
    b.  FilterPatterns is the filter value for log data to be streamed to your logging account (leave empty to stream all logs in the selected log group)
    c.  LogGroupName is the log group as it appears under CloudWatch Logs
  5. Verify successful creation of the subscription

This completes the deployment process in both the logging- and application-account side. After a few minutes, log data will be streamed to the central-logging destination defined in your logging account.

Step 5: Analyzing log data

Once log data is centralized, it opens the door to run analytics on the consolidated data for business or security reasons. One of the powerful services that AWS offers is Amazon Athena.

Amazon Athena allows you to query data in S3 using standard SQL.

Follow the steps below to create a simple table and run queries on the flow logs data that has been collected from your application accounts

  1. Login to your logging account and from the Amazon Athena console, use the DDL below in your query  editor to create a new table

CREATE EXTERNAL TABLE IF NOT EXISTS prod_vpc_flow_logs (

Version INT,

Account STRING,

InterfaceId STRING,

SourceAddress STRING,

DestinationAddress STRING,

SourcePort INT,

DestinationPort INT,

Protocol INT,

Packets INT,

Bytes INT,

StartTime INT,

EndTime INT,

Action STRING,

LogStatus STRING

)

ROW FORMAT SERDE ‘org.apache.hadoop.hive.serde2.RegexSerDe’

WITH SERDEPROPERTIES (

“input.regex” = “^([^ ]+)\\s+([0-9]+)\\s+([^ ]+)\\s+([^ ]+)\\s+([^ ]+)\\s+([^ ]+)\\s+([^ ]+)\\s+([^ ]+)\\s+([^ ]+)\\s+([^ ]+)\\s+([0-9]+)\\s+([0-9]+)\\s+([^ ]+)\\s+([^ ]+)$”)

LOCATION ‘s3://central-logging-company-do-not-delete/’;

2. Click ”run query” and verify a successful run/ This creates the table “prod_vpc_flow_logs”

3. You can then run queries against the table data as below:

Conclusion

By following the steps I’ve outlined, you will build a central logging solution to stream CloudWatch logs from one application account to a central logging account. This solution is repeatable and could be deployed multiple times for multiple accounts and logging requirements.

 

About the Author

Mahmoud Matouk is a Senior Cloud Infrastructure Architect. He works with our customers to help accelerate migration and cloud adoption at the enterprise level.

 

How I built a data warehouse using Amazon Redshift and AWS services in record time

Post Syndicated from Stephen Borg original https://aws.amazon.com/blogs/big-data/how-i-built-a-data-warehouse-using-amazon-redshift-and-aws-services-in-record-time/

This is a customer post by Stephen Borg, the Head of Big Data and BI at Cerberus Technologies.

Cerberus Technologies, in their own words: Cerberus is a company founded in 2017 by a team of visionary iGaming veterans. Our mission is simple – to offer the best tech solutions through a data-driven and a customer-first approach, delivering innovative solutions that go against traditional forms of working and process. This mission is based on the solid foundations of reliability, flexibility and security, and we intend to fundamentally change the way iGaming and other industries interact with technology.

Over the years, I have developed and created a number of data warehouses from scratch. Recently, I built a data warehouse for the iGaming industry single-handedly. To do it, I used the power and flexibility of Amazon Redshift and the wider AWS data management ecosystem. In this post, I explain how I was able to build a robust and scalable data warehouse without the large team of experts typically needed.

In two of my recent projects, I ran into challenges when scaling our data warehouse using on-premises infrastructure. Data was growing at many tens of gigabytes per day, and query performance was suffering. Scaling required major capital investment for hardware and software licenses, and also significant operational costs for maintenance and technical staff to keep it running and performing well. Unfortunately, I couldn’t get the resources needed to scale the infrastructure with data growth, and these projects were abandoned. Thanks to cloud data warehousing, the bottleneck of infrastructure resources, capital expense, and operational costs have been significantly reduced or have totally gone away. There is no more excuse for allowing obstacles of the past to delay delivering timely insights to decision makers, no matter how much data you have.

With Amazon Redshift and AWS, I delivered a cloud data warehouse to the business very quickly, and with a small team: me. I didn’t have to order hardware or software, and I no longer needed to install, configure, tune, or keep up with patches and version updates. Instead, I easily set up a robust data processing pipeline and we were quickly ingesting and analyzing data. Now, my data warehouse team can be extremely lean, and focus more time on bringing in new data and delivering insights. In this post, I show you the AWS services and the architecture that I used.

Handling data feeds

I have several different data sources that provide everything needed to run the business. The data includes activity from our iGaming platform, social media posts, clickstream data, marketing and campaign performance, and customer support engagements.

To handle the diversity of data feeds, I developed abstract integration applications using Docker that run on Amazon EC2 Container Service (Amazon ECS) and feed data to Amazon Kinesis Data Streams. These data streams can be used for real time analytics. In my system, each record in Kinesis is preprocessed by an AWS Lambda function to cleanse and aggregate information. My system then routes it to be stored where I need on Amazon S3 by Amazon Kinesis Data Firehose. Suppose that you used an on-premises architecture to accomplish the same task. A team of data engineers would be required to maintain and monitor a Kafka cluster, develop applications to stream data, and maintain a Hadoop cluster and the infrastructure underneath it for data storage. With my stream processing architecture, there are no servers to manage, no disk drives to replace, and no service monitoring to write.

Setting up a Kinesis stream can be done with a few clicks, and the same for Kinesis Firehose. Firehose can be configured to automatically consume data from a Kinesis Data Stream, and then write compressed data every N minutes to Amazon S3. When I want to process a Kinesis data stream, it’s very easy to set up a Lambda function to be executed on each message received. I can just set a trigger from the AWS Lambda Management Console, as shown following.

I also monitor the duration of function execution using Amazon CloudWatch and AWS X-Ray.

Regardless of the format I receive the data from our partners, I can send it to Kinesis as JSON data using my own formatters. After Firehose writes this to Amazon S3, I have everything in nearly the same structure I received but compressed, encrypted, and optimized for reading.

This data is automatically crawled by AWS Glue and placed into the AWS Glue Data Catalog. This means that I can immediately query the data directly on S3 using Amazon Athena or through Amazon Redshift Spectrum. Previously, I used Amazon EMR and an Amazon RDS–based metastore in Apache Hive for catalog management. Now I can avoid the complexity of maintaining Hive Metastore catalogs. Glue takes care of high availability and the operations side so that I know that end users can always be productive.

Working with Amazon Athena and Amazon Redshift for analysis

I found Amazon Athena extremely useful out of the box for ad hoc analysis. Our engineers (me) use Athena to understand new datasets that we receive and to understand what transformations will be needed for long-term query efficiency.

For our data analysts and data scientists, we’ve selected Amazon Redshift. Amazon Redshift has proven to be the right tool for us over and over again. It easily processes 20+ million transactions per day, regardless of the footprint of the tables and the type of analytics required by the business. Latency is low and query performance expectations have been more than met. We use Redshift Spectrum for long-term data retention, which enables me to extend the analytic power of Amazon Redshift beyond local data to anything stored in S3, and without requiring me to load any data. Redshift Spectrum gives me the freedom to store data where I want, in the format I want, and have it available for processing when I need it.

To load data directly into Amazon Redshift, I use AWS Data Pipeline to orchestrate data workflows. I create Amazon EMR clusters on an intra-day basis, which I can easily adjust to run more or less frequently as needed throughout the day. EMR clusters are used together with Amazon RDS, Apache Spark 2.0, and S3 storage. The data pipeline application loads ETL configurations from Spring RESTful services hosted on AWS Elastic Beanstalk. The application then loads data from S3 into memory, aggregates and cleans the data, and then writes the final version of the data to Amazon Redshift. This data is then ready to use for analysis. Spark on EMR also helps with recommendations and personalization use cases for various business users, and I find this easy to set up and deliver what users want. Finally, business users use Amazon QuickSight for self-service BI to slice, dice, and visualize the data depending on their requirements.

Each AWS service in this architecture plays its part in saving precious time that’s crucial for delivery and getting different departments in the business on board. I found the services easy to set up and use, and all have proven to be highly reliable for our use as our production environments. When the architecture was in place, scaling out was either completely handled by the service, or a matter of a simple API call, and crucially doesn’t require me to change one line of code. Increasing shards for Kinesis can be done in a minute by editing a stream. Increasing capacity for Lambda functions can be accomplished by editing the megabytes allocated for processing, and concurrency is handled automatically. EMR cluster capacity can easily be increased by changing the master and slave node types in Data Pipeline, or by using Auto Scaling. Lastly, RDS and Amazon Redshift can be easily upgraded without any major tasks to be performed by our team (again, me).

In the end, using AWS services including Kinesis, Lambda, Data Pipeline, and Amazon Redshift allows me to keep my team lean and highly productive. I eliminated the cost and delays of capital infrastructure, as well as the late night and weekend calls for support. I can now give maximum value to the business while keeping operational costs down. My team pushed out an agile and highly responsive data warehouse solution in record time and we can handle changing business requirements rapidly, and quickly adapt to new data and new user requests.

Additional Reading

If you found this post useful, be sure to check out Deploy a Data Warehouse Quickly with Amazon Redshift, Amazon RDS for PostgreSQL and Tableau Server and Top 8 Best Practices for High-Performance ETL Processing Using Amazon Redshift.

About the Author

Stephen Borg is the Head of Big Data and BI at Cerberus Technologies. He has a background in platform software engineering, and first became involved in data warehousing using the typical RDBMS, SQL, ETL, and BI tools. He quickly became passionate about providing insight to help others optimize the business and add personalization to products. He is now the Head of Big Data and BI at Cerberus Technologies.