Best-in-class hardware performance with zero-touch accelerated RDMA over Converged Ethernet (RoCE)
NVIDIA is the leader in end-to-end accelerated networking for all layers of software and hardware. NVIDIA combines the benefits of NVIDIA Spectrum™ switches, based on industry-leading application-specific integrated circuit (ASIC) technology, with a wide variety of modern network operating system choices, including NVIDIA Cumulus® Linux and Pure SONiC.
Best-in-class hardware performance with zero-touch accelerated RDMA over Converged Ethernet (RoCE)
Fifth-generation in-house ASIC design for optimized AI, cloud, and storage workloads
Designed and tested for full-stack, end-to-end NVIDIA solutions
Choice and flexibility at all tiers of the stack through open NOS support
Hosted data center digital twins with end-to-end platform simulation
Exclusive features that enable fairness, predictability, and actionable visibility
Purpose-built for AI, the SN5000 family is the world’s only Ethernet switch portfolio suitable for deep learning workloads, connecting cloud-scale GPU compute at speeds up to 800 gigabits per second (Gb/s).
Built for cloud-scale networking, the SN4000 family combines industry-leading features with groundbreaking performance to support modern scale-out distributed data center applications for all speeds up to 400Gb/s.
Ideal for leaf and spine data center network solutions, the SN3000 series delivers maximum flexibility with port speeds up to 200Gb/s per port and port density that enables full-rack connectivity to any server at any speed.
With speeds up to 100Gb/s, the SN2000 series provides unique switch form factors that integrate into hyperconverged Infrastructure and software-defined storage management systems for ease of deployment and administration.
Max Port Speed
800Gb/s
Max Flow Counters
512K entries
Max Access-Control Lists (ACLs)
512K entries
Network Address Translation (NAT)
100K+ entries
Max IPV4 Routes
512K entries
Connectors | 800GbE Ports | 400GbE Ports | 200GbE Ports | 100GbE Ports | 50GbE Ports | 40GbE Ports | 25/10/1GbE Ports | Height | Max Throughput | Total Packets per Second | |
SN5400 | 64x QSFP-DD 400GbE + 2x SFP28 25GbE | 64 | 128 | 256 | 256 | 128 | 256+2 | 2U | 25.6Tb/s | 33.3Bpps | |
SN5600 | 64x OSFP 800GbE + 1 SFP28 25GbE | 64 | 128 | 256 | 256 | 256 | 128 | 256+1 | 2U | 51.2Tb/s | 33.3Bpps |
Connectors | 400GbE Ports | 200GbE Ports | 100GbE Ports | 50GbE Ports | 40GbE Ports | 25/10/1GbE Ports | Height | Max Throughput | Total Packets per Second | |
SN4600C | 64x QSFP28 100GbE | 64 | 128 | 64 | 128 | 2U | 6.4Tb/s | 8.4Bpps | ||
SN4700 | 32x QSFPDD 400GbE | 32 | 64 | 128 | 128 | 64 | 128 | 1U | 12.8Tb/s | 8.4Bpps |
SN4410 | 24x QSFPDD-28 + 8x QSFPDD | 8 | 16 | 24/48 | 128 | 64 | 128 | 1U | 8Tb/s | 8.4Bpps |
SN4600 | 64x QSFP56 200GbE | 64 | 128 | 128 | 64 | 128 | 2U | 12.8Tb/s | 8.4Bpps |
Connectors | 200GbE Ports | 100GbE Ports | 50GbE Ports | 40GbE Ports | 25/10/1GbE Ports | Height | Max Throughput | Total Packets per Second | |
SN3700 | 32x QSFP56 200GbE | 32 | 64 | 128 | 32 | 128 | 1U | 6.4Tb/s | 8.33Bpps |
SN3700C | 32x QSFP28 100GbE | — | 32 | 64 | 32 | 128 | 1U | 3.2Tb/s | 4.76Bpps |
SN3750-SX | 32x QSFP56 200GbE | 32 | 64 | 128 | 32 | 128 | 1U | 6.4Tb/s | 8.33Bpps |
SN3420 | 12x QSFP28 100GbE + 48x SFP28 25GbE |
— | 12 | — | 12 | 48+48 | 1U | 2.4Tb/s | 3.57Bpps |
Connectors | 100GbE Ports | 50GbE Ports | 40GbE Ports | 25/10/1GbE Ports | Height | Max Throughput | Total Packets per Second | |
SN2201 | 48x RJ45 + 4x QSFP28 100 GbE |
4 | 8 | 4 | 16(25/10G), 48(1G) |
1U | 448Gb/s | 667Mpps |
SN2010 | 18x SFP28 25GbE + 4x QSFP28 100GbE |
4 | 8 | 4 | 18+16 | 1U | 850Gb/s | 1.26Bpps |
SN2100 | 16x QSFP28 100GbE | 16 | 32 | 16 | 64 | 1U | 1.6Tb/s | 2.38Bpps |
NVIDIA Air allows users to model data center deployments with full software functionality, creating a digital twin. Transform and streamline network operations by simulating, validating, and automating changes and updates.
NVIDIA Cumulus Linux, the world’s most robust open networking operating system, includes a comprehensive list of advanced, modern networking features and is built for scale.
NVIDIA NetQ™ is a modern operations tool designed to provide holistic, real-time visibility, troubleshooting, and lifecycle management of your modern data center.
NVIDIA offers Pure SONiC, a community-developed, open-source network operating system based on Linux that runs on switches from multiple vendors and powers some of the largest data centers in the world.
NVIDIA Linux Switch lets users natively use any standard Linux distribution as the switch operating system on any of the NVIDIA Spectrum Open Ethernet switch platforms.
Discover the key benefits of NVIDIA’s Ethernet switch solutions highlighted in the IDC Business Value whitepaper
Simulate a fully automated network topology using NVIDIA Air.
Catch these recommended sessions on-demand.
Data Centers have changed. APIs have replaced CLIs, and the scope and scale needed to manage cloud and AI-scale networks make simulation, automation, and validation a necessity. Consisting of Air, Cumulus Linux, and NetQ, the NVIDIA Spectrum Ethernet software stack is purpose-built for operational innovation. Combining network digital twins with powerful, simple automation and actionable visibility, Spectrum Ethernet accelerates Day 0, Day 1, and Day 2 ops for faster time to value and better business outcomes. In this session, learn how to leverage Air, Cumulus, and NetQ for your mission-critical cloud networking.
Modern cloud and telecommunications services require high-performance networks without compromising on openness for their mission-critical infrastructure and applications. Comcast will speak about how they leverage NVIDIA’s open Ethernet networking technologies for their Cloud Fabric services deployment, including Pure SONiC, Spectrum switches, and BlueField DPUs. Learn how to architect and build network topologies based on Spectrum Ethernet switches running the SONiC network operating system for innovative, open 400G networking, ready for the next generation of AI and HPC applications.
Computing is advancing at incredible speed, and cloud computing is becoming the norm for nearly every enterprise. Cloud computing allows companies to deploy applications faster, reduce operating costs, and rapidly scale resources to meet growing demand. However, not all clouds are created equal. Modern applications increasingly rely on accelerated computing as the underlying technology to deliver on the promise of machine learning and artificial intelligence. We'll discuss how organizations can build cloud computing infrastructures that are accelerated, efficient, and secure to power the next wave of applications.