Issue # 141: October 17, 2025 to October 31, 2025

Brought to you by CAST AI​

Hey Reader.

Welcome to Issue # 141 of AWS Graviton Weekly, which will be focused on sharing everything that happened in the past week related to AWS Silicon: from October 15, 2025, to October 31, 2025.

Before sharing the usual stuff we put every week here, we wanted to say two things. The first one is related to the AWS outage. I know for sure this wasn't an easy period for the AWS crew, so HugOps for the whole team behind this.

We were all blaming to DNS (It's always DNS) for the outage, but after reading the post-mortem post, it seems that the actual root cause was a race condition in an internal DynamoDB microservice that automates DNS record management for the regional cells of DynamoDB.

I highly recommend reading the entire post. Believe me: you won't regret it.

The second thing I wanted to address here is the big layoffs that happened at Amazon this week. If you were part of this layoff, please reach out to me directly, and I will share your CV across my network.

Now, back to the regular stuff.

Enjoy

Recommendation of the week: CAST AI: Meet Kelsey Hightower at KubeCon​

NEWS

• ​AWS activates Project Rainier: One of the world’s largest AI compute clusters comes online​
• ​Introducing the Capacity Reservation Topology API for AI, ML, and HPC instance types​
• ​Amazon Aurora DSQL now supports FIPS 140-3 compliant endpoints​
• ​Amazon EC2 High Memory U7i instances are now available in AWS Europe (London) Region​

ARTICLES AND TUTORIALS

• ​Introducing the Arm MCP Server: Simplifying cloud migration with AI, by Zach Lasiuk (Arm)

​

• ​Why your AWS bill secretly hates Graviton, by Fernando Piñero Estrada​

• ​AMD EPYC Turin vs. Intel Xeon 6 Granite Rapids vs. Graviton4 Benchmarks With AWS M8 Instances, by Michael Larabel (Photonix)

• ​How the Python Software Foundation Future-Proofed Its Infrastructure with Arm, by Arm​
  ​
• ​From EC2 to EKS: Inside Coinbase’s 10x Compute Modernization, by Tiberiu Oprisiu, Frances Chong, and Sara Reddy from Coinbase​
  ​

Note: I know for sure they could use CAST AI to be even more efficient with their Kubernetes ops, so I will send them the benefits of using the platform.
​
​

• ​A scientific approach to workload-aware computing on AWS, by Shree Gautam (AWS)

​

• ​Optimizing PyTorch Model Inference on AWS Graviton, by Chaim Rand (Mobileye)

• ​How Smarsh reduced costs and increased scale migrating from Pivotal Cloud Foundry to Amazon EKS, by Krishna Reddy (AWS), Utsab Roy (AWS), Evan Doyle (Smarsh), and Blake Sherwood (Smarsh)

​

• ​emptyvessel builds global game development pipeline with AWS, by Lilly McDermott (AWS), Kevon Mayers (AWS), ​Garrett Young (emptyvessel), and Wei Ning (emptyvessel)

​

• ​Reducing GitHub Actions Costs, by WarpBuild​

VIDEOS

• ​August 2025 Product updates: Kafka & ClickHouse integration, faster Kafka on AWS Graviton & more! by NetApp InstaClustr ​
  ​

From the Cloud Silicon ecosystem

• ​Build smarter and faster: Hands-on Kubernetes and AI workshop on Google Arm-based Axion CPUs​
• ​Arm, Meta Partner to Cut AI Power Costs and Optimize Cloud Performance​​
• ​The Dawn of New Desktop Devices: Arm-Powered NVIDIA DGX Spark Workstations to Redefine AI Computing ​
  ​

Brought to you by CAST AI​

​

See ya next week.

​​

​