Environment:Sgl project Sglang Kubernetes
Sgl_project_Sglang_Kubernetes is the Kubernetes cluster environment for SGLang, enabling orchestrated deployment of SGLang serving pods with GPU scheduling, autoscaling, and multi-node distributed inference.
Requirements
- Kubernetes 1.26+ cluster
- NVIDIA GPU Operator or device plugin for GPU scheduling
- `kubectl` CLI configured with cluster access
- Container images with SGLang and CUDA runtime
- LeaderWorkerSet (LWS) CRD for distributed inference topologies
- Persistent volume claims for model weight caching
- Service and Ingress resources for API endpoint exposure
Required By
See Also
Page Connections
Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment