Switches & Bridges

The SN3000 series switches are the 3rd generation of NVIDIA® Mellanox® Spectrum® switches, purpose-built for leaf/spine/super-spine datacenter applications. Allowing maximum flexibility, SN3000 series provides port speeds spanning from 1GbE to 400GbE, and a port density that enables full rack connectivity to any server at any speed. In addition,the uplink ports allow a variety of blocking ratios to suit any application requirement.

The SN3000 series is ideal for building wire-speed and cloud-scale layer-2 and layer-3 networks. The SN3000 platforms deliver high performance, consistent low latency along with support for advanced software defined networking features, making it the ideal choice for web scale IT, cloud, hyperconverged storage and data analytics applications.

NETWORK DISAGGREGATION: OPEN ETHERNET

Open Ethernet breaks the paradigm of traditional switch systems, eliminating vendor lock-in. Instead of forcing network operators to use the specific software that is provided by the switch vendor, Open Ethernet offers the flexibility to use a choice of operating systems on top of Ethernet switches, thereby re-gaining control of the network, and optimizing utilization, efficiency and overall return on investment.

Encouraging an ecosystem of open source, standard network solution, Open Ethernet adopts the same principles as standard open solutions for servers and storage, and applies them to the world of networking infrastructure. These solutions can then be easily deployed into the modern data center across network equipment that eases management and ensures full interoperability.

With a range of system form factors, and a rich software ecosystem, SN3000 series allows you to pick and choose the right components for your data center.

Visibility

• NVIDIA Mellanox What Just Happened® (WJH) telemetry dramatically reduces mean time to issue resolution by providing answers to: When, What, Who, Where and Why
• Hardware-accelerated histograms track and summarize queue depths at submicrosecond granularity
• Inband Network Telemetry (INT)-ready hardware
• Streaming Telemetry
• 512K on-chip flow counters

• Fully shared packet buffer provides a fair, predictable and high bandwidth data path
• Consistent and low cut-through latency
• Robust RoCE transport to power NVMe over fabric and Machine Learning applications that leverage NVIDIA Mellanox GPUDirect
• Best-in-class VXLAN scale-10X more tunnels and tunnel endpoints than others
• 512K shared forwarding entries flexibly shared across ACL, LPM routes, host routes, MAC, ECMP and tunnel applications

• Comprehensive Layer-2, Layer-3 and RoCE
• Advanced network virtualization with high performance single pass VXLAN routing and IPv6 segment routing
• Cloud Scale NAT - 100K+ sessions
• Programmable pipeline that can programmatically parse, process and edit packets
• Deep Packet Inspection - 512B deep

Mellanox SN3700C is a 1U 32-port 100GbE spine that can also be used as a high density 10/25GbE leaf when used with splitter cables. SN3700C allows for maximum flexibility, with ports spanning from 1GbE to 100GbE and port density that enables full rack connectivity to any server at any speed, and a variety of blocking ratios. SN3700C ports are fully splittable to up to 128 x 10/25GbE ports.

SN3700C Series Switches

The SN3700C is a 1U 32-port 100GbE Open Ethernet Network spine that can also be used as a high density 10/25GbE leaf when used with splitter cables. The SN3700C switch allows for maximum flexibility, with ports spanning from 1GbE to 100GbE and port density that enables full rack connectivity to any server at any speed, and a variety of blocking ratios. SN3700C ports are fully splittable to up to 128X 10/25GbE ports.

Visibility

• NVIDIA What Just Happened (WJH) telemetry dramatically reduces mean time to issue resolution by providing answers to: When, What, Who, Where and Why
• Hardware-accelerated histograms track and summarize queue depths at submicrosecond granularity
• Inband Network Telemetry (INT)-ready hardware
• Streaming Telemetry
• 512K on-chip flow counters

Performance

• Fully shared packet buffer provides a fair, predictable and high bandwidth data path
• Consistent and low cut-through latency
• Robust RoCE transport to power NVMe over fabric and Machine Learning applications that leverage NVIDIA GPUDirect
• Best-in-class VXLAN scale 10X more tunnels and tunnel endpoints than others
• 512K shared forwarding entries flexibly shared across ACL, LPM routes, host routes, MAC, ECMP and tunnel applications

Agility

• Comprehensive Layer-2, Layer-3 and RoCE
• Advanced network virtualization with high performance single pass VXLAN routing and IPv6 segment routing
• Cloud Scale NAT 100K+ sessions
• Programmable pipeline that can programmatically parse, process and edit packets
• Deep Packet Inspection 512B deep

Mellanox SN3700 spine/super-spine offers 32 ports of 200GbE in a compact 1U form factor. It enables connectivity to endpoints at different speeds and carries a throughput of 12.8Tb/s, with a landmark 8.33Bpps processing capacity. As an ideal spine solution, the SN3700 allows maximum flexibility, with port speeds spanning from 10GbE to 200GbE per port.

Mellanox SN3700 spine/super-spine offers 32 ports of 200GbE in a compact 1U form factor. It enables connectivity to endpoints at different speeds and carries a throughput of 12.8Tb/s, with a landmark 8.33Bpps processing capacity. As an ideal spine solution, the SN3700 allows maximum flexibility, with port speeds spanning from 10GbE to 200GbE per port.