-
Vendor:NVIDIA
NVIDIA 920-9N201-00FA-0C0 | MSN3700-VS2FC 32x 200GbE QSFP56, 1U, Linux NOS, P2C airflow
920-9N201-00FA-0C0
Vendor:NVIDIANVIDIA 920-9N201-00FA-0C0 | MSN3700-VS2FC 32x 200GbE QSFP56, 1U, Linux NOS, P2C airflow
High‑density 200GbE leaf/spine switch with 32 QSFP56 ports in a compact 1U form factor. Each port supports breakout to 2x100GbE or 4x50GbE, enabling flexible aggregation and scale-out designs. Built on a Linux-based network OS, it delivers line‑rate L2/L3 switching with EVPN-VXLAN, advanced QoS, and telemetry. Hot‑swappable fans and PSUs with P2C airflow support resilient operations in modern data centers. -
Vendor:NVIDIA
NVIDIA 920-9N201-00F7-0C0 | MSN3700-CS2FC 32x 100GbE QSFP28, license: Linux (Cumulus)
920-9N201-00F7-0C0
Vendor:NVIDIANVIDIA 920-9N201-00F7-0C0 | MSN3700-CS2FC 32x 100GbE QSFP28, license: Linux (Cumulus)
High‑density 1U 100GbE leaf/spine switch delivering wire‑rate performance on 32 QSFP28 ports. Ideal for spine or leaf roles in EVPN‑VXLAN fabrics and HPC/AI clusters requiring predictable, ultra‑low latency and RoCEv2‑friendly QoS. Runs a Linux‑based NOS for feature‑rich L2/L3, automation, and advanced telemetry. -
Vendor:NVIDIA
NVIDIA 920-9N110-00R7-0C2 | MSN2010-CB2RC 18x 10/25GbE SFP28 + 4x 40/100GbE QSFP28, PoE: none, uplinks: 4x 100G, license: Linux (Cumulus)
22‑port leaf/edge switch pairing 18x 10/25G SFP28 access with 4x 40/100G QSFP28 uplinks. Runs a Linux‑based NOS, enabling EVPN‑VXLAN fabrics, robust L2/L3 routing, automation, and deep visibility on Spectrum silicon. Optimized for consistent low latency and lossless Ethernet in cloud and storage networks. -
Vendor:NVIDIA
NVIDIA 920-9N110-00F7-0C3 | SN2010 18x 10/25GbE SFP28 + 4x 40/100GbE QSFP28, PoE: none, uplinks: 4x 100G, license: L2/L3 managed
920-9N110-00F7-0C3
Vendor:NVIDIANVIDIA 920-9N110-00F7-0C3 | SN2010 18x 10/25GbE SFP28 + 4x 40/100GbE QSFP28, PoE: none, uplinks: 4x 100G, license: L2/L3 managed
Versatile 22‑port Spectrum‑based switch combining 18x 10/25G SFP28 access ports with 4x 40/100G QSFP28 uplinks. Delivers line‑rate L2/L3, EVPN‑VXLAN, and precise QoS with ultra‑low latency, ideal for leaf/edge roles bridging 25G servers to 100G spines. Managed L2/L3 feature set provides robust routing, automation hooks, and rich telemetry.
AKA MSN2010-CB2FC -
Vendor:NVIDIA
NVIDIA 920-9N100-00R7-0N0 | SN21000 16x 100GbE QSFP28, License: ONIE (BYOL)
920-9N100-00R7-0N0
Vendor:NVIDIANVIDIA 920-9N100-00R7-0N0 | SN21000 16x 100GbE QSFP28, License: ONIE (BYOL)
Compact, high‑density 100GbE top‑of‑rack switch built on Spectrum silicon for deterministic, ultra‑low latency L2/L3 forwarding. The 16 QSFP28 ports run at 100/40GbE with breakout to 4x 25GbE per port for flexible leaf designs. Ships with ONIE so you can install your preferred NOS (e.g., Cumulus Linux, SONiC). Ideal for EVPN‑VXLAN fabrics, storage (RoCEv2), and cloud workloads requiring lossless QoS and wire‑rate performance.
MSN2100-CB2RO
-
Vendor:NVIDIA
NVIDIA 920-9N100-00R7-0C0 | SN21000 16x 100GbE QSFP28, half‑width 1U, Cumulus Linux (C2P airflow)
920-9N100-00R7-0C0
Vendor:NVIDIANVIDIA 920-9N100-00R7-0C0 | SN21000 16x 100GbE QSFP28, half‑width 1U, Cumulus Linux (C2P airflow)
High‑performance 16‑port 100GbE top‑of‑rack switch with Cumulus Linux, delivering compact density and automated operations for modern data centers. QSFP28 ports provide flexible 25GbE breakouts, while the half‑width 1U chassis simplifies high‑density deployments and dual‑switch side‑by‑side mounting.
MSN2100-CB2RC
-
Vendor:NVIDIA
NVIDIA 920-9N100-00F7-0C0 | SN21000 16x 100GbE QSFP28, half‑width 1U, Cumulus Linux (P2C airflow)
920-9N100-00F7-0C0
Vendor:NVIDIANVIDIA 920-9N100-00F7-0C0 | SN21000 16x 100GbE QSFP28, half‑width 1U, Cumulus Linux (P2C airflow)
Compact 16‑port 100GbE leaf switch for top‑of‑rack deployments, preloaded with Cumulus Linux. The half‑width 1U design delivers high density in constrained racks, while QSFP28 ports support flexible breakouts for 25GbE leaf fan‑out. Ideal for EVPN/VXLAN fabrics, leaf‑spine data centers, and performance‑sensitive workloads.
MSN2100-CB2FC
-
Vendor:NVIDIA
NVIDIA 920-9B110-00FH-0M2 | MTQ8100-HS2F Long-haul extension switch, InfiniBand/Ethernet transport, 10 km reach
920-9B110-00FH-0M2
Vendor:NVIDIANVIDIA 920-9B110-00FH-0M2 | MTQ8100-HS2F Long-haul extension switch, InfiniBand/Ethernet transport, 10 km reach
A compact long-haul extension platform designed to carry high-bandwidth data center fabrics over metro distances. It transparently transports InfiniBand or Ethernet traffic with low latency and deterministic performance, enabling seamless interconnect between sites up to 10 km away. Ideal for extending HPC, AI, and storage networks across campuses or metropolitan areas without re-architecting the fabric. -
Vendor:NVIDIA
NVIDIA 920-9B110-00FH-0M1 | MTQ8200-HS2F 100GbE long-haul switch | ports: QSFP28 100G | uplinks: QSFP28 100G | reach: up to 40 km (with ER4 optics)
Long-haul 100GbE switch engineered for metro and regional data center interconnects. Using QSFP28 optics, it provides deterministic, low-latency transport over distances up to 40 km when paired with suitable ER4-class transceivers, and supports shorter reaches with LR4 optics. Ideal for extending leaf–spine fabrics across campuses or metro sites, it offers L2 transport/retimer functionality with familiar network OS management and monitoring. -
Vendor:NVIDIA
NVIDIA 920-9B110-00FH-0D4 | HDR 200Gb/s InfiniBand chassis module | ports: chassis-integrated | uplinks: backplane HDR 200G | PoE: none | license: UFM-managed
Modular InfiniBand HDR (200Gb/s) component for director-class chassis, delivering high-throughput, low-latency fabric capability for dense HPC and AI clusters. Installed into a compatible HDR chassis, it leverages the backplane for connectivity and is coordinated by the chassis supervisor. UFM provides comprehensive fabric orchestration, telemetry, and health monitoring. -
Vendor:NVIDIA
NVIDIA 920-9B10A-00FH-0D0 | HDR 200Gb/s InfiniBand chassis module | ports: chassis-integrated | uplinks: backplane HDR 200G | PoE: none | license: UFM-managed
Director-class InfiniBand HDR (200Gb/s) chassis component for scaling high-performance fabrics. This module augments capacity and resiliency within an HDR chassis, delivering consistent low latency and high bandwidth for AI and HPC environments. It is administered via the chassis supervisor and UFM, enabling fabric-wide telemetry and lifecycle operations. The host chassis supplies power and cooling. -
Vendor:NVIDIA
NVIDIA 920-9B020-00FH-XD0 | HDR 200Gb/s InfiniBand chassis module | ports: chassis-integrated | uplinks: backplane HDR 200G | PoE: none | license: UFM-managed
High-speed InfiniBand HDR (200Gb/s) chassis component designed for director-class fabrics. This module integrates into an HDR chassis to expand capacity and deliver non-blocking, low-latency connectivity for large-scale HPC and AI clusters. Managed through the chassis supervisor and Unified Fabric Manager (UFM), it supports fabric-wide monitoring, automation, and policy-driven operations. Power and cooling are provided by the host chassis. -
Vendor:NVIDIA
NVIDIA 920-9B020-00FH-0M0 | HDR 200Gb/s InfiniBand | Director-class chassis switch | High-radix core fabric
920-9B020-00FH-0M0
Vendor:NVIDIANVIDIA 920-9B020-00FH-0M0 | HDR 200Gb/s InfiniBand | Director-class chassis switch | High-radix core fabric
High-radix, modular chassis switch for HDR (200Gb/s) InfiniBand deployments. Built to anchor large AI/HPC clusters with consistent, low-latency performance and high bisectional bandwidth. Supports large-scale fabric growth, robust telemetry, and link-level congestion management for predictable performance under mixed workloads. -
Vendor:NVIDIA
NVIDIA 920-9B020-00F0-0D1 | HDR 200Gb/s InfiniBand | Director-class chassis switch | Modular fabric core
920-9B020-00F0-0D1
Vendor:NVIDIANVIDIA 920-9B020-00F0-0D1 | HDR 200Gb/s InfiniBand | Director-class chassis switch | Modular fabric core
Modular director-class chassis platform for HDR (200Gb/s) InfiniBand fabrics, engineered for large AI/HPC clusters requiring predictable latency and extreme bisectional bandwidth. Provides high-radix scalability, congestion control features, and adaptive routing to maintain performance at scale. Ideal as a non-blocking core for compute, storage, and GPU‑dense environments. -
Vendor:NVIDIA
NVIDIA 900-9X6AG-0048-ST0 | 200GbE, QSFP56, PCIe Gen4 x16, FHHL, RoCEv2, PTP/SyncE, inline crypto
900-9X6AG-0048-ST0
Vendor:NVIDIANVIDIA 900-9X6AG-0048-ST0 | 200GbE, QSFP56, PCIe Gen4 x16, FHHL, RoCEv2, PTP/SyncE, inline crypto
High‑performance Ethernet adapter leveraging ConnectX‑6 Dx silicon to deliver up to 200 GbE with advanced hardware offloads. Ideal for scale‑out storage, AI/ML pipelines, and cloud infrastructure requiring ultra‑low latency, congestion control, and precise time synchronization. -
Vendor:NVIDIA
NVIDIA 900-9X6AG-0016-ST0 | 100GbE, QSFP28, PCIe Gen4 x16, FHHL, RoCEv2, PTP/SyncE, inline IPsec/TLS
900-9X6AG-0016-ST0
Vendor:NVIDIANVIDIA 900-9X6AG-0016-ST0 | 100GbE, QSFP28, PCIe Gen4 x16, FHHL, RoCEv2, PTP/SyncE, inline IPsec/TLS
Enterprise Ethernet adapter based on the ConnectX‑6 Dx architecture delivering up to 100 GbE with extensive hardware offloads. Designed for accelerated clouds, NVMe‑oF/TCP or RoCE storage, microservices, and virtualized environments that require line‑rate encryption and precise time synchronization.
-
Why buy NVIDIA from XS Network Tech?
Trusted by data centres and AI infrastructure teams across Australia, XS Network Tech delivers NVIDIA networking hardware with the speed and precision high-throughput environments demand. Our expert sales team supports pre-sales specification for GPU fabrics and AI builds, with local Sydney stock available for same-day dispatch. Wide inventory across current and EOL NVIDIA platforms, every used unit tested and backed by a 12-month warranty.
-
Can I buy used NVIDIA networking hardware in Australia?
Yes. We stock new, used, and EOL NVIDIA networking equipment locally in our Sydney warehouse for fast Australia-wide dispatch. Used NVIDIA hardware is a cost-effective option for teams building out AI infrastructure or high-throughput data centre fabrics without new-hardware lead times or pricing. We also offer configure-to-order NVIDIA GPU servers. Spec your CPU, memory, storage, and networking to match your exact workload, and we'll build and test it before despatch.
-
What NVIDIA products does XS Network Tech stock?
Our NVIDIA range spans ConnectX SmartNICs (ConnectX-6, ConnectX-7), Spectrum Ethernet switches (SN4600, SN4700), DAC and AOC cables (MCP, MFS series), Quantum InfiniBand switches, and GPU compute including RTX A-series, RTX 4000/6000 Ada, and RTX Blackwell, plus BlueField DPUs and Jetson edge AI modules. Can't find your model? Get in touch and we will sort it.
