From tenant-aware to job-aware: scaling shared AI clusters with Cisco Nexus One

AI clusters increasingly require shared infrastructure across multiple tenants and workloads, but current EVPN/VXLAN solutions lack the granularity to distinguish between individual jobs within a tenant. Cisco’s emerging Job-ID-aware segmentation technology extends isolation and visibility beyond tenant boundaries to individual scheduler jobs, optimizing high-performance connectivity for demanding AI and machine learning workloads at scale.