Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

Recommendations Summary

The AI Data Center Network with Juniper Apstra, NVIDIA GPUs, and WEKA Storage JVD follows an industry-standard dedicated IP Fabric design. Three distinct fabrics provide maximum efficiency while maintaining focus on AI model scale, expedited completion times, and rapid evolution with the advent of AI technologies.

To follow best practice recommendations:

  • A minimum of 4 spines in each fabric is suggested.
  • Follow a rail-optimized fabric and maintain a 1:1 subscription factor in the GPU backend fabric.
  • Maintain a 1:1 subscription factor in the Storage backend fabric.
  • Implement Advanced Load Balancing mechanism instead of traditional ECMP for optimal load distribution in the GPU Backend Fabric.
  • Implement DCQCN (PFC and ECN) to ensure a lossless fabric in the GPU Backend Fabric.
  • Configure DCQCN (PFC and ECN) parameters on the AMD servers and change the NCCL_SOCKET interface to be the management (frontend) interface.
  • Configure DCQCN (PFC and ECN) parameters on the Nvidia servers and change the NCCL_SOCKET interface to be the management (frontend) interface.
  • The minimum recommended Junos OS releases for this JVD are:
    • Junos OS Release 23.4R2-S3 is for the Juniper QFX5130-32CD
    • Junos OS Release 23.4X100-D20 for Juniper QFX5220-32CD
    • Junos OS Release 23.4X100-D20 for Juniper QFX5230-64CD
    • Junos OS Release 23.4X100-D31 for Juniper QFX5240-64CD
    • Junos OS Release 23.4X100-D42 for Juniper QFX5240-64OD/QD
    • Junos OS Release 23.4R2-S3 for Juniper PTX10008
    • Apstra 6.1

The Juniper hardware listed in the Juniper Hardware and Software Components section are the best-suited switch platforms regarding features, performance, and the roles specified in this JVD.