Tag Archives: Amazon Kendra

Simplify document search at scale with intelligent search bot on AWS

Post Syndicated from Rostislav Markov original https://aws.amazon.com/blogs/architecture/simplify-document-search-at-scale-with-intelligent-search-bot-on-aws/

Enterprise document management systems (EDMS) manage the lifecycle and distribution of documents. They often rely on keyword-based search functionality. However, it increasingly becomes hard to discover documents as such repositories grow to tens of thousands of items.

In this blog, we discuss how Amazon Web Services (AWS) built an intelligent search bot on top of the document repository of a global life sciences company. Before this enhancement, the native repository search function relied solely on keywords and document names, leading to a trial-and-error process. Now, scientists can effortlessly query the repository using natural language to locate the right document in a few seconds—even through voice commands while working with lab equipment.

Use case

In life sciences, documentation is critical for regulatory compliance with GxP. Scientists in life sciences use EDMS on a daily basis to retrieve standard operating procedures (SOPs). SOPs contain task-level instructions (for example, how to monitor lab equipment or use utilities such as steam generators and chilled water circulation pumps).

EDMS search capability is often limited to file names and metadata. In our use case, file names were numerical and metadata was typically a single-sentence description plus keywords.

Scientists wanted to query for a particular context and type of task they’re about to perform. However, this data was not extracted from document content and thus not readily available for search purposes. Scientists also wanted to be able to search by using a voice interface (for example, when they operate on lab equipment).

To address this, we designed a conversational bot interface. This bot locates the most relevant SOP based on pre-generated document extracts and returns a hyperlink to the most suitable document, as shown in Figure 1.

Example of document search prompt and chatbot response

Figure 1. Example of document search prompt and chatbot response

Overview of the intelligent search bot solution

Our solution requirements for intelligent search were:

  • Semantic search index and ranking based on full text
  • Search capability through voice and text
  • Out-of-the-box integration with web applications and mobile devices

We choose Amazon Lex to provide the conversational interface using text or speech. Lex bots can be integrated with web and mobile applications using AWS Amplify Interactions. We used Amazon Kendra to create an intelligent search index on top of the data repository, which we hosted on Amazon Simple Storage Service (Amazon S3).

The advantage of using an Amazon Kendra index is its out-of-the-box semantic search and ranking capability based on document content. We use AWS Lambda to take care of Amazon S3 path mappings, and document attribute formatting for replicated documents, in order to make them retrievable by Amazon Kendra.

Intelligent search bot for enterprise document management systems

Figure 2. Intelligent search bot for enterprise document management systems

Benefits of integrating intelligent search bot with your EDMS

The benefits of extending EDMS with intelligent search bot include:

  • Improved usability by adding text- and speech-based channels to match user situations (for example, scientists operating on lab equipment)
  • Native, out-of-the-box integration with third-party systems (for example, Adobe Experience Manager, Alfresco, HubSpot, Marketo, Salesforce)
  • Implementation timeboxed in a two-week agile sprint and requires no data science skills

Extensibility to large language models

The Amazon Kendra Retrieve API allows the extension to a document retrieval chain pattern using a large language model (LLM) from Amazon Bedrock or Amazon SageMaker Jumpstart. With this pattern, Amazon Kendra generated document summaries can be passed to the LLM for correlation, as shown in Figure 3. Consult the LangChain documentation to learn how to configure retrieval chains.

Extending the document search bot to large language model

Figure 3. Extending the document search bot to large language model

In our use case, the effort for such extension went beyond the incremental optimizations and limited migration timeframe. Also, preference for a chain pattern should be given to complex correlations across documents.

This wasn’t the case here, as documents were functionally disjunct (for example, by device type, geographical site, and process task). Therefore, document retrieval with the Amazon Kendra API was sufficient and we deferred the extra effort associated with custom-built LLM prompt layers.

Implementation considerations

We started the EDMS migration to AWS by replicating the data repository to Amazon S3 by using AWS DataSync. We stored every document and corresponding metadata files as separate Amazon S3 objects.

For the Amazon Kendra index mappings to initialize properly:

  • Metadata must be a JSON Amazon S3 object
  • Follow the name convention <document>.<extension>.metadata.json
  • Have reserved or common document attributes correctly formatted

The EDMS system did not adhere to this when generating metadata files so we offloaded the transformation to a Lambda function. The function fixed metadata attributes such as, changing version type (_version) from numeric to string and date (_created_at) from string to ISO8601. It also changed metadata names/Amazon S3 paths by enacting new objects (PutObject API) and deleting the original objects (DeleteObject API).

We configured Lambda invocation on Amazon S3 PutObject operations using Amazon S3 event notifications. We set the sync run schedule for the Amazon Kendra index to run on demand.

Alternatively, you can run it on a predefined schedule or as part of each Lambda invocation (using the update_index boto3 operation). Finally, we monitor for sync run fails associated with the Amazon Kendra index using Amazon CloudWatch.


This blog showed how you can enhance the keyword-based search of your EDMS. We embedded document search queries behind a chatbot to simplify user interaction.

You can build this solution quickly, with no machine learning skills, as part of your EDMS cloud migration. In more advanced use cases, including complete refactoring, consider extending this solution to use a large language model from Amazon Bedrock or Amazon SageMaker Jumpstart.

Related information

AWS Week in Review – AWS Documentation Updates, Amazon EventBridge is Faster, and More – May 22, 2023

Post Syndicated from Danilo Poccia original https://aws.amazon.com/blogs/aws/aws-week-in-review-aws-documentation-updates-amazon-eventbridge-is-faster-and-more-may-22-2023/

AWS Data Hero Anahit Pogosova keynote at CloudConf 2023Here are your AWS updates from the previous 7 days. Last week I was in Turin, Italy for CloudConf, a conference I’ve had the pleasure to participate in for the last 10 years. AWS Hero Anahit Pogosova was also there sharing a few serverless tips in front of a full house. Here’s a picture I took from the last row during her keynote.

On Thursday, May 25, I’ll be at the AWS Community Day in Dublin to celebrate the 10 years of the local AWS User Group. Say hi if you’re there!

Last Week’s Launches
Last week was packed with announcements! Here are the launches that got my attention:

Amazon SageMakerGeospatial capabilities are now generally available with security updates and more use case samples.

Amazon DetectiveSimplify the investigation of AWS Security Findings coming from new sources such as AWS IAM Access Analyzer, Amazon Inspector, and Amazon Macie.

Amazon EventBridge – EventBridge now delivers events up to 80% faster than before, as measured by the time an event is ingested to the first invocation attempt. No change is required on your side.

AWS Control Tower – The service has launched 28 new proactive controls that allow you to block non-compliant resources before they are provisioned for services such as AWS OpenSearch Service, AWS Auto Scaling, Amazon SageMaker, Amazon API Gateway, and Amazon Relational Database Service (Amazon RDS). Check out the original posts from when proactive controls were launched.

Amazon CloudFront – CloudFront now supports two new control directives to help improve performance and availability: stale-while-revalidate (to immediately deliver stale responses to users while it revalidates caches in the background) and the stale-if-error cache (to define how long stale responses should be reused if there’s an error).

Amazon Timestream – Timestream now enables to export query results to Amazon S3 in a cost-effective and secure manner using the new UNLOAD statement.

AWS Distro for OpenTelemetryThe tail sampling and the group-by-trace processors are now generally available in the AWS Distro for OpenTelemetry (ADOT) collector. For example, with tail sampling, you can define sampling policies such as “ingest 100% of all error cases and 5% of all success cases.”

AWS DataSync – You can now use DataSync to copy data to and from Amazon S3 compatible storage on AWS Snowball Edge Compute Optimized devices.

AWS Device Farm – Device Farm now supports VPC integration for private devices, for example, when an unreleased version of an app is accessing a staging environment and tests are accessing internal packages only accessible via private networking. Read more at Access your private network from real mobile devices using AWS Device Farm.

Amazon Kendra – Amazon Kendra now helps you search across different content repositories with new connectors for Gmail, Adobe Experience Manager Cloud, Adobe Experience Manager On-Premise, Alfresco PaaS, and Alfresco Enterprise. There is also an updated Microsoft SharePoint connector.

Amazon Omics – Omics now offers pre-built bioinformatic workflows, synchronous upload capability, integration with Amazon EventBridge, and support for Graphical Processing Units (GPUs). For more information, check out New capabilities make it easier for healthcare and life science customers to get started, build applications, and scale-up on Amazon Omics.

Amazon Braket – Braket now supports Aria, IonQ’s largest and highest fidelity publicly available quantum computing device to date. To learn more, read Amazon Braket launches IonQ Aria whith built-in error mitigation.

For a full list of AWS announcements, be sure to keep an eye on the What’s New at AWS page.

Other AWS News
A few more news items and blog posts you might have missed:

AWS Documentation home page screenshot.AWS Documentation – The AWS Documentation home page has been redesigned. Leave your feedback there to let us know what you think or to suggest future improvements. Last week we also announced that we are retiring the AWS Documentation GitHub repo to focus our resources to directly improve the documentation and the website.

Peloton case studyPeloton embraces Amazon Redshift to unlock the power of data during changing times.

Zoom case studyLearn how Zoom implemented streaming log ingestion and efficient GDPR deletes using Apache Hudi on Amazon EMR.

Nice solutionIntroducing an image-to-speech Generative AI application using SageMaker and Hugging Face.

For AWS open-source news and updates, check out the latest newsletter curated by Ricardo to bring you the most recent updates on open-source projects, posts, events, and more.

Upcoming AWS Events
Here are some opportunities to meet and learn:

AWS Data Insights Day (May 24) – A virtual event to discover how to innovate faster and more cost-effectively with data. This event focuses on customer voices, deep-dive sessions, and best practices of Amazon Redshift. You can register here.

AWS Silicon Innovation Day (June 21) – AWS has designed and developed purpose-built silicon specifically for the cloud. Join to learn AWS innovations in custom-designed Amazon EC2 chips built for high performance and scale in the cloud. Register here.

AWS re:Inforce (June 13–14) – You can still register for AWS re:Inforce. This year it is taking place in Anaheim, California.

AWS Global Summits – Sign up for the AWS Summit closest to where you live: Hong Kong (May 23), India (May 25), Amsterdam (June 1), London (June 7), Washington, DC (June 7-8), Toronto (June 14), Madrid (June 15), and Milano (June 22). If you want to meet, I’ll be at the one in London.

AWS Community Days – Join these community-led conferences where event logistics and content is planned, sourced, and delivered by community leaders: Dublin, Ireland (May 25), Shenzhen, China (May 28), Warsaw, Poland (June 1), Chicago, USA (June 15), and Chile (July 1).

That’s all from me for this week. Come back next Monday for another Week in Review!


This post is part of our Week in Review series. Check back each week for a quick roundup of interesting news and announcements from AWS!

AWS Week in Review – March 20, 2023

Post Syndicated from Danilo Poccia original https://aws.amazon.com/blogs/aws/aws-week-in-review-march-20-2023/

This post is part of our Week in Review series. Check back each week for a quick roundup of interesting news and announcements from AWS!

A new week starts, and Spring is almost here! If you’re curious about AWS news from the previous seven days, I got you covered.

Last Week’s Launches
Here are the launches that got my attention last week:

Picture of an S3 bucket and AWS CEO Adam Selipsky.Amazon S3 – Last week there was AWS Pi Day 2023 celebrating 17 years of innovation since Amazon S3 was introduced on March 14, 2006. For the occasion, the team released many new capabilities:

Amazon Linux 2023 – Our new Linux-based operating system is now generally available. Sébastien’s post is full of tips and info.

Application Auto Scaling – Now can use arithmetic operations and mathematical functions to customize the metrics used with Target Tracking policies. You can use it to scale based on your own application-specific metrics. Read how it works with Amazon ECS services.

AWS Data Exchange for Amazon S3 is now generally available – You can now share and find data files directly from S3 buckets, without the need to create or manage copies of the data.

Amazon Neptune – Now offers a graph summary API to help understand important metadata about property graphs (PG) and resource description framework (RDF) graphs. Neptune added support for Slow Query Logs to help identify queries that need performance tuning.

Amazon OpenSearch Service – The team introduced security analytics that provides new threat monitoring, detection, and alerting features. The service now supports OpenSearch version 2.5 that adds several new features such as support for Point in Time Search and improvements to observability and geospatial functionality.

AWS Lake Formation and Apache Hive on Amazon EMR – Introduced fine-grained access controls that allow data administrators to define and enforce fine-grained table and column level security for customers accessing data via Apache Hive running on Amazon EMR.

Amazon EC2 M1 Mac Instances – You can now update guest environments to a specific or the latest macOS version without having to tear down and recreate the existing macOS environments.

AWS Chatbot – Now Integrates With Microsoft Teams to simplify the way you troubleshoot and operate your AWS resources.

Amazon GuardDuty RDS Protection for Amazon Aurora – Now generally available to help profile and monitor access activity to Aurora databases in your AWS account without impacting database performance

AWS Database Migration Service – Now supports validation to ensure that data is migrated accurately to S3 and can now generate an AWS Glue Data Catalog when migrating to S3.

AWS Backup – You can now back up and restore virtual machines running on VMware vSphere 8 and with multiple vNICs.

Amazon Kendra – There are new connectors to index documents and search for information across these new content: Confluence Server, Confluence Cloud, Microsoft SharePoint OnPrem, Microsoft SharePoint Cloud. This post shows how to use the Amazon Kendra connector for Microsoft Teams.

For a full list of AWS announcements, be sure to keep an eye on the What’s New at AWS page.

Other AWS News
A few more blog posts you might have missed:

Example of a geospatial query.Women founders Q&A – We’re talking to six women founders and leaders about how they’re making impacts in their communities, industries, and beyond.

What you missed at that 2023 IMAGINE: Nonprofit conference – Where hundreds of nonprofit leaders, technologists, and innovators gathered to learn and share how AWS can drive a positive impact for people and the planet.

Monitoring load balancers using Amazon CloudWatch anomaly detection alarms – The metrics emitted by load balancers provide crucial and unique insight into service health, service performance, and end-to-end network performance.

Extend geospatial queries in Amazon Athena with user-defined functions (UDFs) and AWS Lambda – Using a solution based on Uber’s Hexagonal Hierarchical Spatial Index (H3) to divide the globe into equally-sized hexagons.

How cities can use transport data to reduce pollution and increase safety – A guest post by Rikesh Shah, outgoing head of open innovation at Transport for London.

For AWS open-source news and updates, here’s the latest newsletter curated by Ricardo to bring you the most recent updates on open-source projects, posts, events, and more.

Upcoming AWS Events
Here are some opportunities to meet:

AWS Public Sector Day 2023 (March 21, London, UK) – An event dedicated to helping public sector organizations use technology to achieve more with less through the current challenging conditions.

Women in Tech at Skills Center Arlington (March 23, VA, USA) – Let’s celebrate the history and legacy of women in tech.

The AWS Summits season is warming up! You can sign up here to know when registration opens in your area.

That’s all from me for this week. Come back next Monday for another Week in Review!


AWS Week in Review – January 16, 2023

Post Syndicated from Antje Barth original https://aws.amazon.com/blogs/aws/aws-week-in-review-january-16-2023/

Today, we celebrate Martin Luther King Jr. Day in the US to honor the late civil rights leader’s life, legacy, and achievements. In this article, Amazon employees share what MLK Day means to them and how diversity makes us stronger.

Coming back to our AWS Week in Review—it’s been a busy week!

Last Week’s Launches
Here are some launches that got my attention during the previous week:

AWS Local Zones in Perth and Santiago now generally available – AWS Local Zones help you run latency-sensitive applications closer to end users. AWS now has a total of 29 Local Zones; 12 outside of the US (Bangkok, Buenos Aires, Copenhagen, Delhi, Hamburg, Helsinki, Kolkata, Muscat, Perth, Santiago, Taipei, and Warsaw) and 17 in the US. See the full list of available and announced AWS Local Zones and learn how to get started.

AWS Local Zones Locations

AWS Clean Rooms now available in preview – During AWS re:Invent this past November, we announced AWS Clean Rooms, a new analytics service that helps companies across industries easily and securely analyze and collaborate on their combined datasets—without sharing or revealing underlying data. You can now start using AWS Clean Rooms (Preview).

Amazon Kendra updates – Amazon Kendra is an intelligent search service powered by machine learning (ML) that helps you search across different content repositories with built-in connectors. With the new Amazon Kendra Intelligent Ranking for self-managed OpenSearch, you can now improve the quality of your OpenSearch search results using Amazon Kendra’s ML-powered semantic ranking technology.

Amazon Kendra also released an Amazon S3 connector with VPC support to index and search documents from Amazon S3 hosted in your VPC, a new Google Drive Connector to index and search documents from Google Drive, a Microsoft Teams Connector to enable Microsoft Teams messaging search, and a Microsoft Exchange Connector to enable email-messaging search.

Amazon Personalize updates – Amazon Personalize helps you improve customer engagement through personalized product and content recommendations. Using the new Trending-Now recipe, you can now generate recommendations for items that are rapidly becoming more popular with your users. Amazon Personalize now also supports tag-based resource authorization. Tags are labels in the form of key-value pairs that can be attached to individual Amazon Personalize resources to manage resources or allocate costs.

Amazon SageMaker Canvas now delivers up to 3x faster ML model training time – SageMaker Canvas is a visual interface that enables business analysts to generate accurate ML predictions on their own—without having to write a single line of code. The accelerated model training times help you prototype and experiment more rapidly, shortening the time to generate predictions and turn data into valuable insights.

For a full list of AWS announcements, be sure to keep an eye on the What’s New at AWS page.

Other AWS News
Here are some additional news items and blog posts that you may find interesting:

AWS open-source news and updates – My colleague Ricardo writes this weekly open-source newsletter in which he highlights new open-source projects, tools, and demos from the AWS Community. Read edition #141 here.

ML model hosting best practices in Amazon SageMaker – This seven-part blog series discusses best practices for ML model hosting in SageMaker to help you identify which hosting design pattern meets your needs best. The blog series also covers advanced concepts such as multi-model endpoints (MME), multi-container endpoints (MCE), serial inference pipelines, and model ensembles. Read part one here.

I would also like to recommend this really interesting Amazon Science article about differential privacy for end-to-end speech recognition. The data used to train AI models is protected by differential privacy (DP), which adds noise during training. In this article, Amazon researchers show how ensembles of teacher models can meet DP constraints while reducing error by more than 26 percent relative to standard DP methods.

Upcoming AWS Events
Check your calendars and sign up for these AWS events:

#BuildOnLiveBuild On AWS Live events are a series of technical streams on twitch.tv/aws that focus on technology topics related to challenges hands-on practitioners face today.

  • Join the Build On Live Weekly show about the cloud, the community, the code, and everything in between, hosted by AWS Developer Advocates. The show streams every Thursday at 09:00 US PT on twitch.tv/aws.
  • Join the new The Big Dev Theory show, co-hosted with AWS partners, discussing various topics such as data and AI, AIOps, integration, and security. The show streams every Tuesday at 08:00 US PT on twitch.tv/aws.

Check the AWS Twitch schedule for all shows.

AWS Community Days – AWS Community Day events are community-led conferences that deliver a peer-to-peer learning experience, providing developers with a venue to acquire AWS knowledge in their preferred way: from one another.

AWS Innovate Data and AI/ML edition – AWS Innovate is a free online event to learn the latest from AWS experts and get step-by-step guidance on using AI/ML to drive fast, efficient, and measurable results.

  • AWS Innovate Data and AI/ML edition for Asia Pacific and Japan is taking place on February 22, 2023. Register here.
  • Registrations for AWS Innovate EMEA (March 9, 2023) and the Americas (March 14, 2023) will open soon. Check the AWS Innovate page for updates.

You can browse all upcoming in-person and virtual events.

That’s all for this week. Check back next Monday for another Week in Review!

— Antje

This post is part of our Week in Review series. Check back each week for a quick roundup of interesting news and announcements from AWS!

Field Notes: How to Boost Your Search Results Using Relevance Tuning with Amazon Kendra

Post Syndicated from Sam Palani original https://aws.amazon.com/blogs/architecture/field-notes-how-to-boost-your-search-results-using-amazon-kendra-relevance-tuning/

One challenge enterprises face when they implement an intelligent search solution for their large data sources, is the ability to quickly provide relevant search results. When working with large data sources, not all features or attributes within your data will be equally relevant to all your users. We want to prioritize identifying and boosting specific attributes for your users to provide the most relevant search results.

Relevance in Amazon Kendra tuning allows you to give a boost to a result in the response when the query includes terms that match the attribute. For example, you might have two similar documents but one is created more recently. A good practice is to boost the relevance for the newer (or earlier) document.

Relevance tuning in Amazon Kendra can be performed manually at the Index level or at the query level. In this blog post, we show how to tune an existing index that is connected to external data sources, and ultimately optimize your internal search results.

Solution overview

We will walk through how you can manually tune your index using boosting techniques to achieve the best results. This enables you to prioritize the results from a specific data source so your users get the most relevant results when they perform searches.

Figure 1. Amazon Kendra setup

Figure 1 illustrates a standard Amazon Kendra setup. An Amazon Kendra index is connected to different Amazon Simple Storage Service (Amazon S3) buckets with multiple data sources.

There are two types of user personas. The first persona is administrators who are responsible for managing the index and performing administrative tasks such as access control, index tuning, and so forth. The second persona is users who access Amazon Kendra either directly or through a custom application that can make API search requests on an Amazon Kendra index. You can use relevance tuning to boost results from one of these data sources to provide a more relevant search result.


This solution requires the following:

If you do not have these prerequisites set up, you might check out Create and query an index that walks you through how to create and query an index in Amazon Kendra.

Furthermore, the AWS services you use in this tutorial are within the AWS Free Tier under a 30-day trial.

Step 1 – Check facet definition

First, review your facet definition and confirm it is facetable, displayable, searchable, and sortable.

In the Amazon Kendra console, select your Amazon Kendra index, then select Facet definition in the Data management panel. Confirm that _data_source_id has all of its attributes checked.

Figure 2. Check facet definition

Step 2 – Review data sources

Next, verify that you have at least two data sources for your Amazon Kendra index.

In the Amazon Kendra console, select your Amazon Kendra index, and then select Data sources in the Data management panel. Confirm that your data sources are correctly synced and available. In our example, data-source-2 is an earlier version and contains unprocessed documents compared to sample-datasource that has newer versions and has more relevant content.

Figure 3. Verify data sources

Step 3 – Perform a regular Amazon Kendra search

Next, we will test a regular search without any relevance tuning. Select Search console, and enter the search term Amazon Kendra VPC. Review your search results.

Figure 4. Regular Amazon Kendra search

In our example search results, the document from the second data source 39_kendra-dg_kendra-dg appears as the third result.

Step 4 – Relevance tuning through boosting

Now we will boost the first data source so documents from the first data source are displayed ahead of the other data sources.

Select data source, and boost the first data source sample-datasource to 8. Press the Save button to save your tuning. Wait several seconds for the changes to propagate.

Figure 5. Boost sample-datasource

Step 5 – Perform the search after boosting

Next, we will test the search with relevance tuning applied. In the search text box enter the search term Amazon Kendra VPC. Review your search results.

Figure 6. Searching after boosting

Notice that the search result no longer contains the document from the second data source.

Cleaning up

To avoid incurring any future charges, remove any index created specifically for this tutorial. In the Amazon Kendra console, select your index. Then select Actions, and choose Delete.

Figure 7. Delete index


In this blog post, we showed you how relevance tuning can be used to produce results ranked by their relevance. We also walked you through an example regarding how to manually perform relevance tuning at the index level in Amazon Kendra to boost your search results.

In addition to relevance tuning at the index level, you can also perform relevance tuning at the query level. Finally, check out the What is Amazon Kendra? and Relevance tuning with Amazon Kendra blog posts to learn more.

Scaling up a Serverless Web Crawler and Search Engine

Post Syndicated from Jack Stevenson original https://aws.amazon.com/blogs/architecture/scaling-up-a-serverless-web-crawler-and-search-engine/


Building a search engine can be a daunting undertaking. You must continually scrape the web and index its content so it can be retrieved quickly in response to a user’s query. The goal is to implement this in a way that avoids infrastructure complexity while remaining elastic. However, the architecture that achieves this is not necessarily obvious. In this blog post, we will describe a serverless search engine that can scale to crawl and index large web pages.

A simple search engine is composed of two main components:

  • A web crawler (or web scraper) to extract and store content from the web
  • An index to answer search queries

Web Crawler

You may have already read “Serverless Architecture for a Web Scraping Solution.” In this post, Dzidas reviews two different serverless architectures for a web scraper on AWS. Using AWS Lambda provides a simple and cost-effective option for crawling a website. However, it comes with a caveat: the Lambda timeout capped crawling time at 15 minutes. You can tackle this limitation and build a serverless web crawler that can scale to crawl larger portions of the web.

A typical web crawler algorithm uses a queue of URLs to visit. It performs the following:

  • It takes a URL off the queue
  • It visits the page at that URL
  • It scrapes any URLs it can find on the page
  • It pushes the ones that it hasn’t visited yet onto the queue
  • It repeats the preceding steps until the URL queue is empty

Even if we parallelize visiting URLs, we may still exceed the 15-minute limit for larger websites.

Breaking Down the Web Crawler Algorithm

AWS Step Functions is a serverless function orchestrator. It enables you to sequence one or more AWS Lambda functions to create a longer running workflow. It’s possible to break down this web crawler algorithm into steps that can be run in individual Lambda functions. The individual steps can then be composed into a state machine, orchestrated by AWS Step Functions.

Here is a possible state machine you can use to implement this web crawler algorithm:

Figure 1: Basic State Machine

Figure 1: Basic State Machine

1. ReadQueuedUrls – reads any non-visited URLs from our queue
2. QueueContainsUrls? – checks whether there are non-visited URLs remaining
3. CrawlPageAndQueueUrls – takes one URL off the queue, visits it, and writes any newly discovered URLs to the queue
4. CompleteCrawl – when there are no URLs in the queue, we’re done!

Each part of the algorithm can now be implemented as a separate Lambda function. Instead of the entire process being bound by the 15-minute timeout, this limit will now only apply to each individual step.

Where you might have previously used an in-memory queue, you now need a URL queue that will persist between steps. One option is to pass the queue around as an input and output of each step. However, you may be bound by the maximum I/O sizes for Step Functions. Instead, you can represent the queue as an Amazon DynamoDB table, which each Lambda function may read from or write to. The queue is only required for the duration of the crawl. So you can create the DynamoDB table at the start of the execution, and delete it once the crawler has finished.

Scaling up

Crawling one page at a time is going to be a bit slow. You can use the Step Functions “Map state” to run the CrawlPageAndQueueUrls to scrape multiple URLs at once. You should be careful not to bombard a website with thousands of parallel requests. Instead, you can take a fixed-size batch of URLs from the queue in the ReadQueuedUrls step.

An important limit to consider when working with Step Functions is the maximum execution history size. You can protect against hitting this limit by following the recommended approach of splitting work across multiple workflow executions. You can do this by checking the total number of URLs visited on each iteration. If this exceeds a threshold, you can spawn a new Step Functions execution to continue crawling.

Step Functions has native support for error handling and retries. You can take advantage of this to make the web crawler more robust to failures.

With these scaling improvements, here’s our final state machine:

Figure 2: Final State Machine

Figure 2: Final State Machine

This includes the same steps as before (1-4), but also two additional steps (5 and 6) responsible for breaking the workflow into multiple state machine executions.

Search Index

Deploying a scalable, efficient, and full-text search engine that provides relevant results can be complex and involve operational overheads. Amazon Kendra is a fully managed service, so there are no servers to provision. This makes it an ideal choice for our use case. Amazon Kendra supports HTML documents. This means you can store the raw HTML from the crawled web pages in Amazon Simple Storage Service (S3). Amazon Kendra will provide a machine learning powered search capability on top, which gives users fast and relevant results for their search queries.

Amazon Kendra does have limits on the number of documents stored and daily queries. However, additional capacity can be added to meet demand through query or document storage bundles.

The CrawlPageAndQueueUrls step writes the content of the web page it visits to S3. It also writes some metadata to help Amazon Kendra rank or present results. After crawling is complete, it can then trigger a data source sync job to ensure that the index stays up to date.

One aspect to be mindful of while employing Amazon Kendra in your solution is its cost model. It is priced per index/hour, which is more favorable for large-scale enterprise usage, than for smaller personal projects. We recommend you take note of the free tier of Amazon Kendra’s Developer Edition before getting started.

Overall Architecture

You can add in one more DynamoDB table to monitor your web crawl history. Here is the architecture for our solution:

Figure 3: Overall Architecture

Figure 3: Overall Architecture

A sample Node.js implementation of this architecture can be found on GitHub.

In this sample, a Lambda layer provides a Chromium binary (via chrome-aws-lambda). It uses Puppeteer to extract content and URLs from visited web pages. Infrastructure is defined using the AWS Cloud Development Kit (CDK), which automates the provisioning of cloud applications through AWS CloudFormation.

The Amazon Kendra component of the example is optional. You can deploy just the serverless web crawler if preferred.


If you use fully managed AWS services, then building a serverless web crawler and search engine isn’t as daunting as it might first seem. We’ve explored ways to run crawler jobs in parallel and scale a web crawler using AWS Step Functions. We’ve utilized Amazon Kendra to return meaningful results for queries of our unstructured crawled content. We achieve all this without the operational overheads of building a search index from scratch. Review the sample code for a deeper dive into how to implement this architecture.