AI Infrastructure
AI infrastructure is the compute, networking, and software used to train and serve models.
Follow training and inference stacks from cluster scheduling and networks to model serving. Coverage focuses on deployment, throughput, reliability, and operating cost.
Latest coverage
Articles
NewsvLLM adds day-zero support for Inkling 2 Small
NewsAWS adds cross-layer GPU diagnostics to SageMaker HyperPod
NewsModal joins gVisor maintainers as project moves to CNCF
NewsMHI and Sakura Internet sign memorandum on Japan sovereign-AI infrastructure
NewsVAST Data says 14 customers made more than $10 billion in commitments
NewsEUROPA adapts a 952B-parameter MoE training setup to JUPITER
NewsPyTorch cancels 2.15.1, targets December 17 for 2.16.0
NewsvLLM adds opt-in FlyDSL MonoKernel decode for five models
NewsUS and Malaysia plan cooperation on sovereign AI cloud
NewsExa reports 47% embedding efficiency gain on AWS AI chips
NewsAWS adds EKS training checks and Slurm hung-job recovery to SageMaker HyperPod
NewsAlberta proposes irrigated-land ban for large data centres
Videos
53:56MLOps for Ads Ranking at Pinteres
Dmytro Spodarets·Nov 3, 2023
36:11Reducing NLP Inference costs through model specialisation
Dmytro Spodarets·Jul 10, 2023
59:31Webinar "dstack – a command-line utility to provision infrastructure for ML workflows"
Dmytro Spodarets·Mar 2, 2023
37:39Webinar "Vertex AI Pipelines infrastructure with Terraform"
Dmytro Spodarets·Mar 2, 2023
0:00Webinar "Deploying DL models with Kubernetes and Kubeflow"
Dmytro Spodarets·Oct 3, 2022
0:00Webinar "Pachyderm in production: lessons learned" (RU)
Dmytro Spodarets·Oct 1, 2021
1:08:33Webinar "From research to product with Hydrosphere" (RU)
Dmytro Spodarets·Sep 24, 2021
55:46Webinar "Re-usable pipelines for ML projects with DVC" (RU)
Dmytro Spodarets·Sep 17, 2021
1:20:54Webinar "Deploying deep learning models with Kubernetes and Kubeflow" (RU)
Dmytro Spodarets·Sep 10, 2021