- 36x EDR 100Gb/s ports in a 1U switch
- Up to 7.2Tb/s aggregate switch throughput
- Ultra-low latency between InfiniBand subnets
- InfiniBand Trade Association (IBTA) specification 1.3 and 1.2.1 compliant
- Quality-of-service enforcement
- 1+1 power supply
Products
-
NVIDIA Spectrum SN4410
SKU: 920-9N312-00FB-0C0More Information- Total ports: 48-port 100GbE + 8x 400GbE
- Throughput: 8Tb/s
- Packet Per Second: 8.4Bpps
- System Memory: 16GB
-
NVIDIA Spectrum SN4600
SKU: MSN4600-VS2FCMore Information- Total ports: 64-port 200GbE
- Throughput: 12.8Tb/s
- Packet Per Second: 8.4Bpps
- System Memory: 16GB
-
NVIDIA Spectrum SN4600C
SKU: MSN4600-VS2FC-1More Information- Total ports: 64-port 100GbE
- Throughput: 6.4Tb/s
- Packet Per Second: 8.4Bpps
- System Memory: 8GB
-
NVIDIA Spectrum SN4700
SKU: MSN4700-WS2FMore Information- Total ports: 32-port 400GbE
- Throughput: 12.8Tb/s
- Packet Per Second: 8.4Bpps
- System Memory: 16GB
-
NVIDIA Spectrum SN5400
SKU: 920-9N42C-00RB-7N0More Information- Total ports: 64-port 400GbE + 2x 25GbE
- Switching Capacity: 25.6Tb/s
- Wire-Speed Switching: 33.3Bpps
-
NVIDIA Spectrum SN5600
SKU: 920-9N42F-00RI-7N0More Information- Total ports: 64-port 800GbE + 1x 25GbE
- Switching Capacity: 51.2Tb/s
- Wire-Speed Switching: 33.3Bpps
-
SB7780/7880 InfiniBand Series
SKU: MSB7880-ES2RMore InformationKey Features
Benefits
-
NVIDIA QUADRO GV100
SKU: N/AMore Information- GPU Memory: 32GB HBM2
- CUDA Cores: 5120
- Single Precision: 14.8 TFLOPs
- Double Precision: 7.4 TFLOPs
-
AMD EPYC™ 7H12
SKU: N/AMore Information- # of Cores: 64
- # of Threads: 128
- Processor Base Frequency: 2.60 GHz
- TDP: 280W
-
Intel® Xeon® Platinum 8253
SKU: N/AMore Information- # of Cores: 16
- # of Threads: 32
- Processor Base Frequency: 2.20 GHz
- Cache: 22 MB
-
Intel® Xeon® Platinum 8256
SKU: N/AMore Information- # of Cores: 4
- # of Threads: 8
- Processor Base Frequency: 3.80 GHz
- Cache: 16.5 MB
-
Intel® Xeon® Platinum 8260
SKU: N/AMore Information- # of Cores: 24
- # of Threads: 48
- Processor Base Frequency: 2.40 GHz
- Cache: 35.75 MB
-
Intel® Xeon® Platinum 8268
SKU: N/AMore Information- # of Cores: 24
- # of Threads: 48
- Processor Base Frequency: 2.90 GHz
- Cache: 35.75 MB
-
Intel® Xeon® Platinum 8270
SKU: N/AMore Information- # of Cores: 26
- # of Threads: 52
- Processor Base Frequency: 2.70 GHz
- Cache: 35.75 MB
-
Intel® Xeon® Platinum 8276
SKU: N/AMore Information- # of Cores: 28
- # of Threads: 56
- Processor Base Frequency: 2.20 GHz
- Cache: 38.5 MB
-
Intel® Xeon® Platinum 8280
SKU: N/AMore Information- # of Cores: 28
- # of Threads: 56
- Processor Base Frequency: 2.70 GHz
- Cache: 38.5 MB
-
NVIDIA DGX-2
SKU: N/AMore Information- WORLD’s FIRST 2 PETAFLOPS SYSTEM
- ENTERPRISE GRADE AI INFRASTRUCTURE
- 12 TOTAL NVSWITCHES + 8 EDR INFINIBAND/100 GbE ETHERNET
- 2 INTEL PLATINUM CPUS + 1.5 TB SYSTEM MEMORY + DUAL 10/25 GbE ETHERNET
- 30 TB NVME SSDS INTERNAL STORAGE
-
NVIDIA DGX-1
SKU: N/AMore Information- EFFORTLESS PRODUCTIVITY
- NVIDIA TESLA V100 + NEXT GENERATION NVIDIA NVLINK
- TWO INTEl XEON CPUs + QUAD EDR IB
- THREE-RACK-UNIT ENCLOSURE
-
Blender 3.0 alpha USD branch
SKU: N/AThis is an experimental branch of Blender 3.0 alpha. Specifically, the Universal Scene Description branch which includes the following: Improved USD Import/Export Omniverse MDL import/export through the Universal Material Mapper addon Blender is the free and open source 3D creation suite. It supports the entirety of the 3D pipeline — modeling, rigging, animation, simulation, rendering, ...More Information -
Caffe
SKU: N/ACaffe is a deep learning framework made with expression, speed, and modularity in mind. It is developed by Berkeley AI Research (BAIR) and by community contributors. Yangqing Jia created the project during his PhD at UC Berkeley. Caffe is released under the BSD 2-Clause license. Expressive architecture encourages application and innovation. Models and optimization are defined by configuration without ...More Information -
CUDA
SKU: N/ACUDA is a parallel computing platform and programming model developed by Nvidia for general computing on its own GPUs (graphics processing units). CUDA enables developers to speed up compute-intensive applications by harnessing the power of GPUs for the parallelizable part of the computation. This post is a super simple introduction to CUDA, the popular parallel computing ...More Information -
cuDNN
SKU: N/ANVIDIA cuDNN The NVIDIA CUDA® Deep Neural Network library (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. cuDNN provides highly tuned implementations for standard routines such as forward and backward convolution, pooling, normalization, and activation layers. Deep learning researchers and framework developers worldwide rely on cuDNN for high-performance GPU acceleration. It allows them ...More Information -
DOCKER
SKU: N/AMore InformationDocker is the de facto developer standard for building and sharing apps that enable simplicity, agility and choice for software development across any infrastructure so that you can get your job done and deploy your applications faster. Docker provides developer-friendly, CLI-based workflow and makes it easy to build, share, and run containerized applications. Even your most complex applications can be containerized. You can build locally, deploy to the cloud, and run anywhere. -
Nvidia AI Enterprise NVAIE
SKU: N/AUnleashing AI for Every Enterprise Artificial intelligence is transforming every industry, whether it’s by improvingcustomer relationships in financial services, streamlining manufacturer supplychains, or helping doctors deliver better outcomes for patients. IDC projects that, by2024, 60 percent of the Global 20001 will expand the use of AI and machine learningacross all business-critical horizontal functions, such as ...More Information -
NVIDIA DGX Cloud
SKU: N/ANVIDIA DGX CloudYour own AI supercomputer—in the cloud. Large language models (LLMs) and generative AI require an AI supercomputer, butmany enterprises struggle with the complexity, effort, and time required to deployand operate infrastructure. These businesses need immediate access to high-performance infrastructure at scale, with enterprise-grade tools and predictablecosts that aren’t easily obtainable with traditional ...More Information -
NVIDIA DGX Station
SKU: DGXS-2511C+P2CMI00More Information- Four NVIDIA TESLA V100 GPU
- Next Generation NVIDIA NVLINK
- Water Cooling
- 1/20 Power CONSUMPTION
- Pre-installed standard Ubuntu 14.04 w/ Caffe, Torch, Theano, BIDMach, cuDNN v2, and CUDA 8.0
-
Omniverse Audio2Face
SKU: N/ANVIDIA’s Audio2Face is an Omniverse application that uses a combination of AI technologies to generate facial animation and dialogue lip-sync from an audio source input. The application provides an array of pre- and post-process parameters to fine-tune the animation performance before exporting the result as a geometry cache. Audio2Face requires Windows 64-bit 1909 or above.More Information -
Omniverse Cache
SKU: N/ANVIDIA Omniverse Cache is a service that optimizes data transfers between Nucleus and Omniverse Apps and Connectors on user workstations. Host data with Nucleus Docker and locally cache downloaded data from Nucleus to significantly reduce time to load assets into Omniverse Apps and Connectors.More Information -
Omniverse Create
SKU: N/AOmniverse Create is an application built on Omniverse Kit that accelerates advanced scene composition and allows users to interactively assemble, light, simulate, and render scenes in Pixar USD in real-time. Create lets you easily navigate, modify, and render Pixar USD content in a live, synchronous environment. FEATURES: Simplify Scene Composition: Create offers advanced layout tools ...More Information -
Omniverse Drive
SKU: N/ANVIDIA Omniverse Drive software mounts the Omniverse server to a user’s Windows 10 OS as a virtual hard-drive, giving users the ability to manipulate server assets the same way as local files. Users can copy, read, or write files directly in Windows Explorer or write programs to access files. Omniverse Drive’s capabilities are useful for ...More Information
Our Customers
Previous
Next