581 Search Results for "NVIDIA ConnectX-6"
-
Vendor:NVIDIA
NVIDIA 980-9IAH1-00XM00 | 1600Gb/s OSFP twin-port transceiver, InfiniBand, 1310nm SMF 500m, 2xMPO
980-9IAH1-00XM00
Stock Status: Enquire
Vendor:NVIDIANVIDIA 980-9IAH1-00XM00 | 1600Gb/s OSFP twin-port transceiver, InfiniBand, 1310nm SMF 500m, 2xMPO
High-capacity OSFP twin-port optical transceiver delivering up to 1.6 Tb/s aggregate throughput over single-mode fiber. Uses 1310 nm EML technology to achieve reaches up to 500 m with dual MPO connectors, ideal for structured HPC/AI fabrics and leaf–spine links. -
Vendor:NVIDIA
NVIDIA 980-9I96P-00H030 | 200G QSFP56 AOC | 850nm MMF | 30m | 2x QSFP56 | IB HDR/200GbE
980-9I96P-00H030
Stock Status: Enquire
Vendor:NVIDIANVIDIA 980-9I96P-00H030 | 200G QSFP56 AOC | 850nm MMF | 30m | 2x QSFP56 | IB HDR/200GbE
Long-reach 200G active optical cable built for InfiniBand HDR and 200GbE deployments. QSFP56-to-QSFP56 integrated optics reduce complexity and improve airflow versus discrete modules. Operates at 850nm over multimode fiber with robust PAM4 signaling to maintain performance across 30 meters. -
Vendor:Cisco
Cisco HX-B-GPU-P6-R | GRID P6 vGPU mezz (rear) | blade GPU accelerator | virtualization-ready
HX-B-GPU-P6-R
Stock Status: Enquire
Vendor:CiscoCisco HX-B-GPU-P6-R | GRID P6 vGPU mezz (rear) | blade GPU accelerator | virtualization-ready
Rear-position GPU mezzanine module based on the GRID P6 series for VDI, rendering, and compute acceleration in blade servers. Designed for virtual GPU deployments with compatible software and platforms. -
Vendor:NVIDIA
NVIDIA MCX755106AC-HEAT | 200GbE/NDR200 IB, 2x QSFP112, HHHL, PCIe 5.0 x16 (+x16 ext), HW crypto
MCX755106AC-HEAT
Stock Status: Enquire
Vendor:NVIDIANVIDIA MCX755106AC-HEAT | 200GbE/NDR200 IB, 2x QSFP112, HHHL, PCIe 5.0 x16 (+x16 ext), HW crypto
Half-height, half-length PCIe adapter with dual QSFP112 ports for 2x200GbE or NDR200 InfiniBand, delivering ultra‑low latency and high throughput for AI/HPC clusters. Includes PCIe Gen5 x16 with optional x16 extension, hardware crypto offloads, RDMA/RoCE, NVMe-oF, and precise PTP timing. -
Vendor:NVIDIA
NVIDIA 900-9X658-0046-SB0 | 1x QSFP28 100GbE | PCIe 4.0 x8 | Low-profile | RDMA RoCE v2, SR-IOV, OVS/DPDK offload, PTP
900-9X658-0046-SB0
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-9X658-0046-SB0 | 1x QSFP28 100GbE | PCIe 4.0 x8 | Low-profile | RDMA RoCE v2, SR-IOV, OVS/DPDK offload, PTP
Enterprise-class 100GbE adapter delivering consistent low latency and high throughput for cloud, storage, and NFV workloads. The QSFP28 interface supports 100GbE with flexible transceiver and cable options. Hardware accelerations including RoCEv2, SR-IOV, and overlay offloads minimize CPU overhead. PCIe 4.0 x8 connectivity, precise PTP time stamping, and telemetry features support deterministic data center performance. -
Vendor:Cisco
Cisco HX-GPU-M60 | Tesla M60 Dual-GPU Accelerator, 16GB GDDR5, 300W
HX-GPU-M60
Stock Status: Enquire
Vendor:CiscoCisco HX-GPU-M60 | Tesla M60 Dual-GPU Accelerator, 16GB GDDR5, 300W
The Cisco HX-GPU-M60 is an NVIDIA Tesla M60 GPU accelerator designed for VDI and virtual workstation deployments in HyperFlex and UCS rack servers. Featuring dual Maxwell-architecture GPUs with 4096 CUDA cores total, 16GB GDDR5 memory, and NVIDIA GRID software support, it delivers high-density graphics virtualization for up to 32 concurrent users. Ideal for data centers requiring professional graphics acceleration and virtual desktop infrastructure. -
Vendor:NVIDIA
NVIDIA 980-9I96L-00H020 | 200G QSFP56 AOC | 850nm MMF | 20m | 2x QSFP56 | IB HDR/200GbE
980-9I96L-00H020
Stock Status: Enquire
Vendor:NVIDIANVIDIA 980-9I96L-00H020 | 200G QSFP56 AOC | 850nm MMF | 20m | 2x QSFP56 | IB HDR/200GbE
High-performance 200G active optical cable for low-latency InfiniBand HDR and 200GbE links. Integrated QSFP56 optics on both ends simplify deployment and eliminate separate transceivers. Multimode 850nm design delivers consistent signal integrity across 20 meters, ideal for top-of-rack to aggregation runs. Engineered for low power and minimal BER with PAM4 signaling. -
Vendor:NVIDIA
NVIDIA 980-9I39E-00H001 | 200G QSFP56 DAC breakout, 1m, 1x QSFP56 to 2x QSFP56
980-9I39E-00H001
Stock Status: Enquire
Vendor:NVIDIANVIDIA 980-9I39E-00H001 | 200G QSFP56 DAC breakout, 1m, 1x QSFP56 to 2x QSFP56
High‑performance 1 m passive copper DAC breakout cable for in‑rack or adjacent‑rack connections, mapping a 200G QSFP56 port to two independent 100G QSFP56 links with near‑zero latency and minimal power draw. -
Vendor:NVIDIA
NVIDIA 920-9N42F-00RI-7C0 | Spectrum-4 64-Port 800G Switch, 51.2Tbps, 2U
920-9N42F-00RI-7C0
Stock Status: Enquire
Vendor:NVIDIANVIDIA 920-9N42F-00RI-7C0 | Spectrum-4 64-Port 800G Switch, 51.2Tbps, 2U
The NVIDIA SN5600 is a Spectrum-4 based data center switch delivering 64 ports of 800GbE in a compact 2U form factor with 51.2Tbps aggregate switching capacity. Designed for AI, cloud, and hyperscale environments, it supports spine, super-spine, and leaf deployments with flexible port speeds from 1GbE to 800GbE. This open Ethernet switch ships with Cumulus Linux authentication and dual 3kW AC power supplies with connector-to-power airflow. -
Vendor:NVIDIA
NVIDIA 900-9X625-0053-SB0 | 2x SFP28 10/25GbE | PCIe 4.0 x8 | Low-profile PCIe | RDMA RoCE v2, SR-IOV, VXLAN/NVGRE offload
900-9X625-0053-SB0
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-9X625-0053-SB0 | 2x SFP28 10/25GbE | PCIe 4.0 x8 | Low-profile PCIe | RDMA RoCE v2, SR-IOV, VXLAN/NVGRE offload
High-density 10/25GbE adapter for modern servers, delivering low latency and efficient CPU offload for virtualization, cloud, and storage fabrics. Dual SFP28 cages support 10 or 25GbE with flexible DAC, AOC, or optical modules. Hardware offloads for tunneling and virtualization streamline overlay networks while RoCEv2 enables fast east-west traffic for NVMe/TCP and storage clusters. Designed for compact builds with a low-profile bracket and backward compatibility to PCIe 3.0. -
Vendor:Cisco
Cisco HCI-FI-6454-M6 | Fabric Interconnect (M6 bundle alignment) | up to 54x 10/25G SFP28 | 40/100G QSFP28 uplinks | Non-PoE | License-free switching
HCI-FI-6454-M6
Stock Status: Enquire
Vendor:CiscoCisco HCI-FI-6454-M6 | Fabric Interconnect (M6 bundle alignment) | up to 54x 10/25G SFP28 | 40/100G QSFP28 uplinks | Non-PoE | License-free switching
Compact, M6-aligned fabric interconnect providing 10/25GbE access for servers and 40/100GbE uplinks for aggregation. Ideal for M6-generation hyperconverged and compute bundles, deployed as an HA pair with centralized, policy-based management. -
Vendor:Cisco
Cisco UCSC-GPU-FLEX140 | Intel Flex 140 GPU, 12GB GDDR6, PCIe 4.0 x8, 75W
UCSC-GPU-FLEX140
Stock Status: Enquire
Vendor:CiscoCisco UCSC-GPU-FLEX140 | Intel Flex 140 GPU, 12GB GDDR6, PCIe 4.0 x8, 75W
The Cisco UCSC-GPU-FLEX140 is an Intel Data Center GPU Flex 140 graphics card designed for virtualized desktop infrastructure, media processing, and AI visual inference workloads in data center and edge environments. Featuring 12GB GDDR6 memory, 16 Xe cores, and up to 62 SR-IOV virtual functions with no licensing fees, this half-height, half-length PCIe 4.0 x8 card delivers efficient GPU acceleration in a compact 75W TDP form factor. Ideal for knowledge worker VDI, cloud gaming, media streaming, and light AI inference applications in Cisco UCS C-Series and X-Series servers. -
Vendor:NVIDIA
NVIDIA 900-13701-0040-000 | Jetson AGX Orin AI Module, 32GB LPDDR5, 200 TOPS, 8-Core Arm CPU
900-13701-0040-000
In stock
Vendor:NVIDIANVIDIA 900-13701-0040-000 | Jetson AGX Orin AI Module, 32GB LPDDR5, 200 TOPS, 8-Core Arm CPU
The NVIDIA Jetson AGX Orin 32GB is a compact system-on-module designed for edge AI and autonomous machine applications. Delivering up to 200 TOPS of AI performance with configurable power from 15W to 40W, this module features an 8-core Arm Cortex-A78AE CPU, 1792-core Ampere GPU with 56 Tensor cores, and 32GB LPDDR5 memory. Ideal for robotics, industrial automation, vision AI, and complex multi-sensor fusion deployments requiring high-performance inference at the edge. -
Vendor:NVIDIA
NVIDIA 920-9N110-00F1-0C0 | 48x 1G RJ45 access + 4x 100G QSFP28 uplinks | Non‑PoE | 1U | Cumulus Linux
920-9N110-00F1-0C0
Stock Status: Enquire
Vendor:NVIDIANVIDIA 920-9N110-00F1-0C0 | 48x 1G RJ45 access + 4x 100G QSFP28 uplinks | Non‑PoE | 1U | Cumulus Linux
Open Ethernet 1G/100G switch for campus/edge aggregation and server access. Provides 48x 10/100/1000BASE‑T access ports with 4x 100GbE QSFP28 uplinks, powered by NVIDIA Spectrum switching and an x86 control plane, shipping with Cumulus Linux for modern L2/L3/EVPN fabrics. Dual hot‑swap PSUs for resilience. -
Vendor:NVIDIA
NVIDIA 900-9X760-0018-MB2 | Nvidia MCX753436MC-HEAB Internal Fiber 200000 Mbit/s
900-9X760-0018-MB2
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-9X760-0018-MB2 | Nvidia MCX753436MC-HEAB Internal Fiber 200000 Mbit/s
All products sold by XS Network Tech include a 12-month warranty on both new and used items. Our in-house technical team thoroughly tests used hardware prior to sale to ensure enterprise-grade reliability. -
Vendor:Cisco
Cisco UCSC-GPU-A16-D | A16 Quad-GPU Graphics Card, 64GB GDDR6, VDI Optimized
UCSC-GPU-A16-D
Stock Status: Enquire
Vendor:CiscoCisco UCSC-GPU-A16-D | A16 Quad-GPU Graphics Card, 64GB GDDR6, VDI Optimized
The Cisco UCSC-GPU-A16-D is a professional GPU accelerator designed for high-density virtual desktop infrastructure deployments. Built on NVIDIA Ampere architecture, this quad-GPU board features four 16GB GDDR6 memory partitions totaling 64GB, delivering exceptional performance for VDI, virtual workstations, and graphics-rich remote work environments. Optimized for use with NVIDIA Virtual PC and RTX Virtual Workstation software, it doubles user density compared to previous generations while maintaining superior visual quality. -
Vendor:NVIDIA
NVIDIA 980-9I17Q-00CM00 | 100GbE | QSFP28 | 100GBASE-CWDM4 | 1310 nm | 2 km SMF | LC duplex
980-9I17Q-00CM00
Stock Status: Enquire
Vendor:NVIDIANVIDIA 980-9I17Q-00CM00 | 100GbE | QSFP28 | 100GBASE-CWDM4 | 1310 nm | 2 km SMF | LC duplex
Cost‑effective 100G CWDM4 optics for campus and data center interconnect up to 2 km. QSFP28 with LC duplex uses four CWDM lanes near 1310 nm over OS2 single‑mode fiber for simple, standards‑based deployment. -
Vendor:Cisco
Cisco UCSC-GPU-L4M6 | L4 24GB AI Inference GPU, 72W, Single-Slot PCIe 4.0
UCSC-GPU-L4M6
Stock Status: Enquire
Vendor:CiscoCisco UCSC-GPU-L4M6 | L4 24GB AI Inference GPU, 72W, Single-Slot PCIe 4.0
The Cisco UCSC-GPU-L4M6 is an NVIDIA L4 Tensor Core GPU designed for energy-efficient AI inference, video processing, and virtual workstation deployments in Cisco UCS C-series rack servers. Built on the Ada Lovelace architecture with 24GB GDDR6 ECC memory and 72W TDP, this single-slot, low-profile GPU delivers up to 485 TOPS INT8 performance while drawing power entirely from the PCIe slot without auxiliary cables. Ideal for edge computing, AI inference at scale, video transcoding, and virtualized graphics workloads where rack space and power efficiency are critical. -
Vendor:NVIDIA
NVIDIA 900-5G190-2570-000 | RTX 4000 Ada Graphics Card, 20GB GDDR6, 6144 CUDA Cores
900-5G190-2570-000
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-5G190-2570-000 | RTX 4000 Ada Graphics Card, 20GB GDDR6, 6144 CUDA Cores
The NVIDIA RTX 4000 Ada Generation is a powerful single-slot professional graphics card built on the Ada Lovelace architecture. Designed for workstation environments, it delivers exceptional performance for AI inference, real-time 3D rendering, complex visualization, and GPU-accelerated compute workloads. With 20GB of ECC GDDR6 memory and 6144 CUDA cores, this full-height PCIe 4.0 card excels in professional applications from CAD and engineering simulation to video production and data science. -
Vendor:Cisco
Cisco HCI-GPU-T4-16 | NVIDIA T4 GPU, 16GB GDDR6, 75W, PCIe Gen3 x16
HCI-GPU-T4-16
Stock Status: Enquire
Vendor:CiscoCisco HCI-GPU-T4-16 | NVIDIA T4 GPU, 16GB GDDR6, 75W, PCIe Gen3 x16
The Cisco NVIDIA T4 GPU delivers powerful AI inference and machine learning acceleration in a compact, energy-efficient design. Built on NVIDIA Turing architecture with 2,560 CUDA cores and 320 Tensor Cores, this 16GB GDDR6 graphics processor excels at deep learning inference, video transcoding, virtual desktop infrastructure, and data analytics workloads. Its low-profile 75W PCIe Gen3 x16 form factor enables high-density server deployments while maintaining exceptional performance for enterprise AI applications. -
Vendor:Cisco
Cisco UCSX-GPU-L40 | NVIDIA L40 GPU, 48GB GDDR6, 300W, PCIe Gen4 x16
UCSX-GPU-L40
Stock Status: Enquire
Vendor:CiscoCisco UCSX-GPU-L40 | NVIDIA L40 GPU, 48GB GDDR6, 300W, PCIe Gen4 x16
The Cisco UCSX-GPU-L40 is a full-height, full-length dual-slot GPU accelerator designed for Cisco UCS X-Series modular systems. Built on the NVIDIA Ada Lovelace architecture with 48GB GDDR6 ECC memory and 18,176 CUDA cores, it delivers 90.5 TFLOPS FP32 performance for AI inference, virtual workstations, 3D rendering, and graphics-intensive workloads. The 300W passive-cooled card includes PCIe Gen4 x16 connectivity and supports NVIDIA vGPU for multi-tenant virtualization in enterprise data center environments. -
Vendor:NVIDIA
NVIDIA 900-13701-0050-000 | Jetson AGX Orin Module, 64GB, 275 TOPS, 12-Core CPU
900-13701-0050-000
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-13701-0050-000 | Jetson AGX Orin Module, 64GB, 275 TOPS, 12-Core CPU
The NVIDIA Jetson AGX Orin 64GB Module is a high-performance system-on-module delivering up to 275 TOPS of AI performance for robotics, autonomous machines, and edge AI applications. Featuring a 2048-core Ampere GPU, 12-core Arm Cortex-A78AE CPU, and 64GB LPDDR5 memory, this compact 100x87mm module supports multiple concurrent AI pipelines with power configurable from 15W to 60W. Ideal for demanding workloads including natural language processing, 3D perception, multi-sensor fusion, and vision AI. -
Vendor:Cisco
Cisco UCSC-GPU-A10 | NVIDIA A10 24GB GDDR6 GPU, 150W Passive, PCIe 4.0
UCSC-GPU-A10
Stock Status: Enquire
Vendor:CiscoCisco UCSC-GPU-A10 | NVIDIA A10 24GB GDDR6 GPU, 150W Passive, PCIe 4.0
The Cisco UCSC-GPU-A10 is a compact, single-slot NVIDIA A10 Tensor Core GPU designed for data center inference, virtual desktop infrastructure, and AI workloads. Built on NVIDIA Ampere architecture with 24GB of GDDR6 memory and 150W passive cooling, this PCIe Gen 4 accelerator delivers efficient performance for virtualized graphics, rendering, and compute-intensive applications in Cisco UCS rack servers. -
Vendor:NVIDIA
NVIDIA 900-5G132-2560-000 | RTX 4500 Ada Graphics Card, 24GB GDDR6 ECC, 210W
900-5G132-2560-000
Stock Status: Enquire
Vendor:NVIDIANVIDIA 900-5G132-2560-000 | RTX 4500 Ada Graphics Card, 24GB GDDR6 ECC, 210W
The NVIDIA RTX 4500 Ada Generation is a professional graphics card built on the Ada Lovelace architecture, designed for demanding creative, design, engineering, and scientific workflows. Featuring 7,680 CUDA cores, 60 third-generation RT Cores, and 240 fourth-generation Tensor Cores with 24GB of ECC GDDR6 memory, it delivers exceptional performance for AI-accelerated graphics, real-time ray tracing, and rendering. With 432 GB/s memory bandwidth and PCIe 4.0 x16 interface, the RTX 4500 Ada provides the power and memory capacity needed for complex 3D modeling, simulation, generative AI, and large dataset processing in professional desktop workstations.
