In-Band Network Telemetry: Cisco Silicon One Smart Stamping & Congestion Signaling for AI Networks
1:47
Cisco Silicon One offers in-band telemetry with smart stamping, congestion signaling, and packet trimming. Learn how real-time data embedded in packets enables dynamic congestion control, reduces latency, and optimizes AI workload performance.
Related Videos
In AI Networking
-
Play video Unified Architecture: Single SDK & P4 Programmability Across All Cisco Silicon One Chip Families
Unified Architecture: Single SDK & P4 Programmability Across All Cisco Silicon One Chip Families
Advanced SerDes technology in Cisco Silicon One reduces AI data center power consumption. Discover how integrated retimers enable Linear Pluggable Optics, saving 500W per switch, along with direct copper cables that deliver zero-watt connectivity.
1:44
-
Play video Adaptive Load Balancing: Cisco Silicon One Manages Mixed AI Traffic with Congestion-Aware Routing
Adaptive Load Balancing: Cisco Silicon One Manages Mixed AI Traffic with Congestion-Aware Routing
Cisco Silicon One dynamically manages mixed traffic in distributed AI clusters. Discover microsecond-level congestion detection, intelligent rerouting, and adaptive load balancing that optimizes packet spraying and flowlet-based traffic flows.
1:42
-
Play video Network Programmability: Cisco Silicon One P4-Powered Dynamic ASIC Updates Without Hardware Changes
Network Programmability: Cisco Silicon One P4-Powered Dynamic ASIC Updates Without Hardware Changes
Cisco Silicon One uses P4-powered programmability to enable dynamic ASIC updates without hardware changes. Learn how packet inspection, traffic shaping, advanced routing, and real-time policy enforcement help networks adapt to evolving standards.
1:39
-
Play video Intelligent Load Balancing: Cisco Silicon One Hardware-Driven Fault Recovery for AI Networks
Intelligent Load Balancing: Cisco Silicon One Hardware-Driven Fault Recovery for AI Networks
Cisco Silicon One revolutionizes network recovery with hardware-driven fault detection and instant rerouting. Discover microsecond-level convergence, automatic bandwidth adjustment, and intelligent load balancing for AI workload efficiency.
1:47
-
Play video Reliable Burst Management: Cisco Silicon One Shared Memory Architecture Prevents Packet Loss in AI
Reliable Burst Management: Cisco Silicon One Shared Memory Architecture Prevents Packet Loss in AI
The shared memory architecture of Cisco Silicon One handles synchronized GPU traffic bursts. Learn how dynamic buffer allocation prevents packet drops, ensures uninterrupted data delivery, and improves resource utilization for AI workloads.
1:12
-
Play video Traffic Engineering with Cisco Silicon One: AI-Driven Segment Routing, Policy & Static Pinning
Traffic Engineering with Cisco Silicon One: AI-Driven Segment Routing, Policy & Static Pinning
Explore advanced traffic engineering capabilities enabled by Cisco Silicon One, including AI-driven segment routing, policy-based routing, and static pinning. Optimize data flows and ensure predictable performance.
2:19
