This course was developed by members of AWS Technical Field Communities (TFC), an AWS community of technical experts. The content is intended to complement our standard training curriculum and augment your AWS learning journey. We are aware some courses have accessibility limitations and are working to address. If you require accommodation, please contact AWS Training and Certification Customer Support.
Learners will gain the ability to build a GPU cost attribution solution for Amazon EKS using Amazon Managed Service for Prometheus, Amazon Managed Grafana, and OpenTelemetry. They will learn to achieve granular visibility into GPU resource consumption across shared clusters -- enabling accurate chargeback, waste identification, and capacity planning. The course covers leveraging NVIDIA Multi-Instance GPU (MIG) technology for multi-tenancy and combining AWS managed services with open-source tools to answer critical questions about who is using GPU resources, what it costs, and where inefficiencies exist.
