-
Information
- Fast Ethernet switch with 300 ns port-to-port latency
- High-bandwidth performance with consistent low latency
- Fair bandwidth sharing with fully shared packet buffer
- Rich data center features at unprecedented scale with NVIDIA FlexFlow packet processing
- Advanced network virtualization with support for high-performance VXLAN routing and IPv6 and multi-protocol label switching (MPLS) segment routing
- Programmable pipeline with the ability to programmatically parse, process, and edit packets
- Robust RoCE transport to power NVMe-oF and machine learning applications that leverage GPUDdirect
-
Specifikation
-
Downloadbare filer