Switches & Bridges

The Edgecore AS4610 series is a high-performance Gigabit Ethernet Layer 3 switch family featuring 54 ports; 48 10/100/1000BASE-T ports, 4 x 10G SFP+ uplink ports, and 2 x 20G QSFP+ stacking ports (Stacking capability subject to NOS implementations).

The AS4610-54T is ideal as a data center top-of-rack switch or IPMI (Intelligent Platform Management Interface) switch, it includes redundant hot-swappable AC PSUs and port-to-power or power-to-port (AS4610-54T v1 only) airflow.The AS4610-54P v1 fully supports IEEE 802.3at and IEEE 802.3af Power-over-Ethernet (PoE). The AS4610-54P v1 is an ideal for campus networking and able to provide up to 30 Watts of power per port to attached devices (eight ports support UPoE up to 60 W per port), such as VoIP phones, wireless access points, surveillance cameras etc, fully utilizing existing Cat. 5 cable infrastructure.

The NVIDIA Spectrum SN5000 series switches are the 5th generation of Spectrum switches, purpose-built to accelerate data center fabrics. The SN5000 series provides port speeds spanning from 40GbE to 800GbE, and delivers accelerated Ethernet to every data center. The SN5000 series is ideal for enabling cloud-scale infrastructure for data centers of any size. The SN5000 switch systems provide high performance and consistent low latency along with support for advanced software defined networking features, making them the ideal choice for AI, cloud, data analytics, and simulation applications.

Powered by NVIDIA Spectrum-4 ASICs, the SN5000 series features dynamic, flexible shared buffers and predictable wire speed performance. The SN5000 is built to accelerate NVIDIA platforms, including NVIDIA EGXTM, NVIDIA DGX PODTM , and NVIDIA OVXTM SuperPODs, and AI solution stacks such as NVIDIA AI Enterprise and NVIDIA LaunchPad. As part of the Spectrum platform, the SN5000 systems are pre-tested and pre-validated with NVIDIA's full portfolio of Ethernet networking technology, including BlueField DPUs, ConnectX SmartNICs, and LinkX interconnects. This end-to-end switch to host solution is critical to powering accelerated workloads, and delivers the high performance and innovative feature set needed to supercharge cloud-native applications at scale.

SN5000 switch systems are based on the high-performance Spectrum-4 ASIC with a bidirectional switching capacity of up to 51.2 Tbps. SN5000 platforms are available in a range of configurations, each delivering high performance combined with feature-rich layer 2 and layer 3 forwarding, ideally suited for both top-of-rack leaf and fixed configuration spines. SN5000 series provides full wire speed, cut through-mode latency, on-chip fully-shared 128MB packet buffering, and flexible port use in addition to advanced capabilities. Combining a wide range of innovations in the area of programmability, telemetry, and tunneling with industry-leading performance, NVIDIA SN5000 series is capable of addressing the complex networking requirements of today's data centers.

The SN5600 smart-leaf, spine, and super-spine switch offers 64 ports of 800GbE in a dense 2U form factor. The SN5600 is ideal for NVIDIA Spectrum-X deployments and enables both standard leaf/spine designs with top-of-rack (ToR) switches as well as end-of-row (EoR) topologies. The SN5600 offers diverse connectivity in combinations of 1 to 800GbE and boasts an industry-leading total throughput of 51.2Tb/s.

Accelerate Research, Innovation, and Product Development with Greater Data Capacity and In-Network Computing

As high-performance computing (HPC) and artificial intelligence (AI) applications become more complex, the demand for the most advanced high-speed networking is critical for extreme-scale systems. NVIDIA Quantum-2 is the industry-leading switch platform in power and density, with NDR 400 gigabit per second (Gb/s) InfiniBand throughput that provides AI developers and scientific researchers with the highest networking performance available to take on the world's most challenging problems.

Advanced Computing Needs Advanced Networking

The NVIDIA Quantum-2-based QM9700 and QM9790 switch systems deliver an unprecedented 64 ports of NDR 400Gb/s InfiniBand per port in a 1U standard chassis design. A single switch carries an aggregated bidirectional throughput of 51.2 terabits per second (Tb/s), with a landmark of more than 66.5 billion packets per second (BPPS) capacity. Supporting the latest NDR technology, NVIDIA Quantum-2 brings a highspeed, extremely low-latency and scalable solution that incorporates state-of-the-art technologies such as Remote Direct Memory Access (RDMA), adaptive routing, and NVIDIA Scalable Hierarchical Aggregation and Reduction Protocol (SHARP)

Unlike any other networking solution, NVIDIA InfiniBand provides self-healing network capabilities, as well as quality of service (QoS), enhanced virtual lane (VL) mapping, and congestion control to provide the highest overall application throughput. As an ideal rackmounted InfiniBand solution, the QM9700 and QM9790 NDR InfiniBand fixed-configuration switches allow maximum flexibility, as they enable a variety of topologies, including Fat Tree, SlimFly, DragonFly+, multi-dimensional Torus, and more. They're also backwards compatible to previous generations and include expansive software ecosystem support.

The Era of Data-Driven Computing Today's complex research demands ultra-fast processing of high-resolution simulations, extreme-size datasets, and complex, highly parallelized algorithms that need to exchange information in real time. The QM9700 NDR InfiniBand switches extend NVIDIA In-Network Computing technologies and introduce the third generation of NVIDIA SHARP technology, SHARPv3. Creating virtually unlimited scalability for large data aggregation through the network, SHARPv3 enables support for up to 64 parallel flows-32X higher AI acceleration power compared to the previous generation. SHARPv3 dramatically boosts application performance of complex computations while data moves through the data center network, participating in the application's runtime and reducing the amount of data needed to traverse the network.

NVIDIA MQM9790-NS2R Quantum 2 NDR InfiniBand Switch 64-ports 400Gb/s 32 OSFP Ports Unmanaged C2P Airflow (Reverse)

Scaling Out Data Centers with 400G InfiniBand Smart Switches

The NVIDIA Quantum-2-based QM9700 and QM9790 switch systems deliver an unprecedented 64 ports of 400Gb/s InfiniBand per port in a 1U standard chassis design. A single switch carries an aggregated bidirectional throughput of 51.2 terabits per second (Tb/s), with a landmark of more than 66.5 billion packets per second (BPPS) capacity. Supporting the latest NVIDIA high-speed interconnect 400Gb/s technology, NVIDIA Quantum-2 brings a high-speed, extremely low-latency and scalable solution that incorporates state-of-the-art technologies such as Remote Direct Memory Access (RDMA), adaptive routing, and NVIDIA Scalable Hierarchical Aggregation and Reduction Protocol (SHARP)

The SN2000 switch series is comprised of ONIE (Open Network Install Environment) based platforms which support the mounting of a multitude of operating systems and utilize the advantages of Open Networking and the Mellanox Spectrum™ ASIC capabilities.

Half-Width 16-port Non-blocking 100GbE Open Ethernet Switch System

The SN2100 switch provides a high density, side-by-side 100GbE switching solution which scales up to 64 25GbE ports in 1RU for the growing demands of today's database, storage, data centers environments.

The SN2100 switch is an ideal spine and top of rack (ToR) solution, allowing maximum flexibility, with port speeds spanning from 10Gb/s to 100 Gb/s per port and port density that enables full rack connectivity to any server at any speed. The uplink ports allow a variety of blocking ratios that suit any application requirement.

Powered by the Spectrum ASIC and packed with 16 ports running at 100GbE, the SN2100 carries a whopping switching capacity of 3.2Tb/s with a landmark 4.8Bpps processing capacity in a compact 1RU form factor.

Half-Width 16-port Non-blocking 100GbE Open Ethernet Switch System

The SN2100 switch provides a high density, side-by-side 100GbE switching solution which scales up to 64 25GbE ports in 1RU for the growing demands of today's database, storage, data centers environments.

The SN2100 switch is an ideal spine and top of rack (ToR) solution, allowing maximum flexibility, with port speeds spanning from 10Gb/s to 100 Gb/s per port and port density that enables full rack connectivity to any server at any speed. The uplink ports allow a variety of blocking ratios that suit any application requirement.

Powered by the Spectrum ASIC and packed with 16 ports running at 100GbE, the SN2100 carries a whopping switching capacity of 3.2Tb/s with a landmark 4.8Bpps processing capacity in a compact 1RU form factor.

The SN2100 switch provides a high density, side-by-side 100GbE switching solution which scales up to 128 ports in 1RU for the growing demands of today's database, storage, data centers environments.

The SN2100 switch is an ideal spine and top of rack (ToR) solution, allowing maximum flexibility, with port speeds spanning from 10Gb/s to 100Gb/s per port and port density that enables full rack connectivity to any server at any speed. The uplink ports allow a variety of blocking ratios that suit any application requirement.

Powered by the Mellanox Spectrum™ ASIC and packed with 16 ports running at 100GbE, the SN2100 carries a whopping throughput of 3.2Tb/s with a landmark 4.76Bpps processing capacity in a compact 1RU form factor.

Following the footsteps of the SwitchX®-2 based-systems, the SN2100 enjoys the legacy of the fieldproven MLNX-OS® operating system with a wide installed base and a robust implementation of data, control and management planes that drives the world's most powerful data centers.

Keeping with the Mellanox tradition of setting performance record switch systems, the SN2100 introduces the world's lowest latency for a 100GbE switching and routing element, and does so while having the lowest power consumption in the market. With the SN2100, the use of 25, 40, 50 and 100GbE in large scale is enabled without changing power infrastructure facilities.

The SN2100 is part of Mellanox's complete end-to-end solution which provides 10GbE through 100GbE interconnectivity within the data center. Other devices in this solution include ConnectX®-4 based network interface cards, and LinkX™ copper or fiber cabling. This end-to-end solution is topped with Mellanox NEO™, a management application that relieves some of the major obstacles when deploying a network. NEO enables a fully certified and interoperable design, speeds up time to service and eventually speeds up RoI.

The SN2100 carries a unique design to accommodate the highest rack performance. Its design allows side-by-side placement of two switches in a single, 1RU slot of a 19" rack, delivering high availability to the hosts.

Database solutions require high availability and the ability to scale out in active-active configuration. For example, DB2 pureScale or Oracle RAC require high bandwidth and low latency to the caching facility, the disk storage system, etc., with connectivity to the application servers. The SN2100 is the best fit, providing the highest network throughput, resilience and a mix of 25GbE and 100GbE ports.

The SN2100 introduces hardware capabilities for multiple tunneling protocols that enable increased reachability and scalability for today's data centers. Implementing MPLS, NVGRE and VXLAN tunneling encapsulations in the network layer of the data center allows more flexibility for terminating a tunnel by the network, in addition to termination on the server endpoint.

While Spectrum provides the thrust and acceleration that powers the SN2100, an integrated powerful x86-based processor allows this system to not only be the highest performing switch fabric element, but also gives the ability to incorporate a Linux running server into the same device. This opens up multiple application aspects of utilizing the high CPU processing power and the best switching fabric, to create a powerful machine with unique appliance capabilities that can improve numerous network implementation paradigms.