All posts by Patrick Kennedy

Welcome Back Intel Xeon 6900P Reasserts Intel Server Leadership

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/welcome-back-intel-xeon-6900p-reasserts-intel-server-leadership/

The Intel Xeon 6900P offers twice the cores, huge memory bandwidth and more connectivity. Welcome back Intel to the top end server CPU market

The post Welcome Back Intel Xeon 6900P Reasserts Intel Server Leadership appeared first on ServeTheHome.

MiTAC Intel Denali Pass High-End Liquid Cooled 2U 4-Node Review

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/mitac-intel-denali-pass-high-end-liquid-cooled-2u-4-node-review-coolit/

The MiTAC Intel Denali Pass platform can liquid cool almost everything down to the PSUs. We show the awesome engineering in our review

The post MiTAC Intel Denali Pass High-End Liquid Cooled 2U 4-Node Review appeared first on ServeTheHome.

Tyan GC70-B8033 1U AMD EPYC Milan Server Review

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/tyan-gc70-b8033-bs8033g70e2hr-c-1u-amd-epyc-milan-review-mitac-kioxia-intel-marvell/

In our Tyan GC70-B8033 review, we find a really neatly designed AMD EPYC Milan 1U server designed for easy service with unique features

The post Tyan GC70-B8033 1U AMD EPYC Milan Server Review appeared first on ServeTheHome.

Supermicro ARS-111GL-NHR NVIDIA GH200 Grace Hopper 1U Server Review

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/supermicro-ars-111gl-nhr-nvidia-gh200-grace-hopper-1u-server-review-arm/

We review the Supermicro ARS-111GL-NHR, a 1U air-cooled NVIDIA GH200 server, and show why Grace Hopper is super cool

The post Supermicro ARS-111GL-NHR NVIDIA GH200 Grace Hopper 1U Server Review appeared first on ServeTheHome.

A Quick Introduction to the NVIDIA GH200 aka Grace Hopper

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/a-quick-introduction-to-the-nvidia-gh200-aka-grace-hopper-arm/

The NVIDIA GH200 or “Grace Hopper” is far from a single product. We have a quick guide so when someone says “GH200” you know what to look for

The post A Quick Introduction to the NVIDIA GH200 aka Grace Hopper appeared first on ServeTheHome.

MLPerf Inference v4.1 NVIDIA B200 Whallops AMD MI300X UntetherAI Rises

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/mlperf-inference-v4-1-nvidia-b200-whallops-amd-mi300x-untetherai-rises/

In MLPerf Inference v4.1 the upcoming NVIDIA B200 whalloped the current AMD MI300X but UntetherAI submitted some excellent efficiency runs

The post MLPerf Inference v4.1 NVIDIA B200 Whallops AMD MI300X UntetherAI Rises appeared first on ServeTheHome.

XiangShan High-Performance RISC-V Processors at Hot Chips 2024

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/xiangshan-high-performance-risc-v-processors-at-hot-chips-2024/

The XiangShan RISC-V CPU project from Chinese universities is designing and delivering cores targeting Arm Neoverse N2 and Cortex A76 designs

The post XiangShan High-Performance RISC-V Processors at Hot Chips 2024 appeared first on ServeTheHome.

Cerebras Enters AI Inference Blows Away Tiny NVIDIA H100 GPUs by Besting HBM

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/cerebras-enters-ai-inference-blows-away-tiny-nvidia-h100-gpus-by-besting-hbm/

Cerebras enters AI inference and uses its giant chip to run circles around NVIDIA HGX H100 platform performance

The post Cerebras Enters AI Inference Blows Away Tiny NVIDIA H100 GPUs by Besting HBM appeared first on ServeTheHome.

Intel 4Tbps Optical Chiplet for XPU to XPU Connectivity Detailed

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/intel-4tbps-optical-chiplet-for-xpu-to-xpu-connectivity-detailed/

At Hot Chips 2024 Intel showed off a 4Tbps optical chiplet to displace copper chip-to-chip links with longer reach efficient optics

The post Intel 4Tbps Optical Chiplet for XPU to XPU Connectivity Detailed appeared first on ServeTheHome.

Tesla DOJO Exa-Scale Lossy AI Network using the Tesla Transport Protocol over Ethernet TTPoE

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/tesla-dojo-exa-scale-lossy-ai-network-using-the-tesla-transport-protocol-over-ethernet-ttpoe/

Tesla brought its Dojo V1 networking hardware to Hot Chips 2024 and announced that it is donating its own TTPoE protocol to the Ultra Ethernet Consortium

The post Tesla DOJO Exa-Scale Lossy AI Network using the Tesla Transport Protocol over Ethernet TTPoE appeared first on ServeTheHome.

Meta AI Acceleration in the Next-Gen Meta MTIA for Recommendation Inference

Post Syndicated from Patrick Kennedy original https://www.servethehome.com/meta-ai-acceleration-in-the-next-gen-meta-mtia-for-recommendation-inference-risc-v/

The next-gen Meta MTIA is a custom RISC-V accelerator for the company’s recommendation model AI inference workloads deployed this year

The post Meta AI Acceleration in the Next-Gen Meta MTIA for Recommendation Inference appeared first on ServeTheHome.