DraNet
  • Home
  • User Guides
  • Concepts
  • Contributing
  • DraNet
    • Quick Start
    • User Guides
      • Ray on GKE using DraNet
      • GKE with NVIDIA DRA and DraNet
      • GKE and Cloud TPU v6e (Trillium)
      • GKE and GPUDirect RDMA with DRA
      • MPI Operator on GKE and GPUDirect RDMA
      • Interface Configuration
    • Concepts
      • Linux Network Namespaces and Interfaces
      • Making Networks Flexible
      • Interface Status
      • Hardware Efficiency
      • RDMA
      • RDMA Device Handling
      • How It Works
      • References
    • Contributing
      • Developer Guide
      • Contributing
View page source Edit this page Create child page Create documentation issue
  1. DraNet
  2. Concepts
  3. References

References

  • The Kubernetes Network Driver Model: A Composable Architecture for High-Performance Networking - This paper introduces the Kubernetes Network Driver model and provides a detailed performance evaluation of DraNet, demonstrating significant bandwidth improvements for AI/ML workloads.
  • The Challenges of AI/ML Multi-Node Workloads in Kubernetes - Antonio Ojea, Google - Regular SIG Network Meeting for 2025-07-17
  • Kubernetes Network Drivers, Antonio Ojea, Presentation
  • KEP 3063 - Dynamic Resource Allocation #306

  • KEP 3695 - DRA: structured parameters #438

  • Extend PodResources to include resources from Dynamic Resource Allocation (DRA)

  • Working Group Device Management

  • The Future of Kubernetes Networking - Antonio Ojea, Googe & Dan Winship, Red Hat - Kubernetes Contributor Summit EU 2024

  • Better Together! GPU, TPU and NIC Topological Alignment with DRA - John Belamaric, Google & Patrick Ohly, Intel - Kubecon US 2024

© 2024–2025 Google LLC | CC BY 4.0 |All Rights ReservedPrivacy Policy