https://awslabs.github.io/awsome-distributed-ai/posts/Recent content in Posts on Awsome-Distributed-AI Deep Dives by the AWS Frameworks TeamHugoen-usSun, 17 May 2026 00:00:00 +0000https://awslabs.github.io/awsome-distributed-ai/posts/slurm-qos-on-parallelcluster/Sun, 17 May 2026 00:00:00 +0000https://awslabs.github.io/awsome-distributed-ai/posts/slurm-qos-on-parallelcluster/Use Slurm QoS — associations, limits, fairshare, preemption, partition QoS — to govern shared GPU capacity across teams on top of the multi-user ParallelCluster from Part 1.https://awslabs.github.io/awsome-distributed-ai/posts/slurm-accounting-multi-user-on-parallelcluster/Sat, 16 May 2026 00:00:00 +0000https://awslabs.github.io/awsome-distributed-ai/posts/slurm-accounting-multi-user-on-parallelcluster/Stand up a Slurm accounting database on Aurora Serverless v2, wire it into AWS ParallelCluster, and onboard multiple users without LDAP. Part 1 of a two-post series; Part 2 adds QoS on top.https://awslabs.github.io/awsome-distributed-ai/posts/building-blocks-fm-training-aws/Thu, 07 May 2026 00:00:00 +0000https://awslabs.github.io/awsome-distributed-ai/posts/building-blocks-fm-training-aws/A four-layer view of the foundation-model lifecycle on AWS — accelerator compute and EFA networking, resource orchestration with Slurm and Kubernetes, the ML software stack, and observability.