OcNOS 6.6 Generally Available: Flex-Algo Network Slicing, AI Fabric Foundations, and SR Enhancements

OcNOS 6.6 is now generally available. This release marks the arrival of production-ready Flex-Algo for service provider networks, enabling true network slicing over a single IS-IS domain without additional hardware. For data center operators, OcNOS 6.6 lays the AI fabric groundwork with PFC and ETS support for RoCEv2 workloads, fully delivered in OcNOS 6.6.1 and 7.0.

Flex-Algo: Production Network Slicing

OcNOS 6.6 delivers full Flex-Algo production support including delay-metric and TE-metric algorithm types, per-algorithm SID assignment, and SR-TE policy integration. Network operators can now define multiple logical topologies, one optimized for latency-sensitive 5G traffic and another for high-bandwidth bulk data, over a single physical IS-IS domain without additional hardware or overlay complexity.

! OcNOS 6.6: Production Flex-Algo deployment
!
router isis CORE
  ! Algorithm 128: optimize for minimum link delay (5G, real-time)
  flex-algo 128
    metric-type link-delay
    advertise
  !
  ! Algorithm 129: optimize for TE metric (bulk bandwidth)
  flex-algo 129
    metric-type te-metric
    advertise
  !
  segment-routing mpls
!
! Per-algorithm SIDs on loopback
interface lo
  prefix-sid index 1           ! Algo 0 (default)
  prefix-sid algorithm-num 128 index 1001  ! Algo 128
  prefix-sid algorithm-num 129 index 2001  ! Algo 129
!
! Verify Flex-Algo topology:
show isis flex-algo
show isis segment-routing state | include algo

OcNOS 6.6 for Data Centers

OcNOS 6.6 DC introduces initial PFC (Priority-Based Flow Control) and ETS (Enhanced Transmission Selection) support for Broadcom-based platforms, the lossless fabric foundations required for RoCEv2 AI workloads. Full Tomahawk 5 AI fabric support arrives in OcNOS 6.6.1 and 7.0.

New Platforms in OcNOS 6.6

• OcNOS 6.6 for Service Providers
• OcNOS 6.6 for Data Centers
• OcNOS 6.6.1 Release
• Flex-Algo Technical Guide
• Contatta IP Infusion

IP Infusion Product Team