A technical guide to Kubernetes DRA for GPUs. Learn how to define hardware requirements via ResourceClaims and optimize costs using GPU sharing (MPS, MIG).

Cast AI is a platform offering insights, tutorials, and resources for cloud infrastructure and Kubernetes users. Readers can learn about cloud-native technologies, container orchestration, and infrastructure optimization. With tutorials, best practices, and case studies, Cast AI helps organizations optimize their cloud resources and streamline their Kubernetes deployments.

Cast AI

Kubernetes 1.34 made Dynamic Resource Allocation (DRA) generally available, replacing the blunt `nvidia.com/gpu: 1` device plugin model with structured, attribute-based GPU requests. DRA introduces four core objects: ResourceSlice (describes available hardware), DeviceClass (groups devices), ResourceClaimTemplate (per-pod GPU requirements), and ResourceClaim (shared across pods). A practical walkthrough shows deploying a CUDA Mandelbrot fractal renderer with precise GPU requirements (Ampere architecture, 20+ GB memory), then scaling it using three GPU sharing strategies: time-slicing (sequential access), MPS (concurrent CUDA processes with configurable thread percentages), and MIG (hardware-level partitioning into up to seven isolated 1g.5gb slices on an A100). CAST AI's autoscaler integrates natively with DRA, reading ResourceClaims to provision the cheapest matching instance type automatically, including setting up MIG partitions.

Deploying GPU workload with Dynamic Resource Allocation

Deploying with DRA with precise GPU requirements