- Rack Height: 4U
- Processor: 2x Intel Xeon Scalable family
- Drive Bays: 6x 2.5" SATA/NVMe Hot-Swap
- 8x NVIDIA A100 SXM4 GPUs
The TensorEX TS4-168747704-DPN is a 2U rack mountable Deep Learning & AI server supporting 2x Intel Xeon Scalable family processors, a maximum of 8 TB DDR4 memory, and eight A100 Ampere GPUs (SXM4), with up to 600GB/s NVLINK interconnect.
GPUs have provided groundbreaking performance to accelerate deep learning research with thousands of computational cores and up to 100x application throughput when compared to CPUs alone. Exxact has developed the TensorEX TS4-168747704-DPN, featuring NVIDIA GPU technology coupled with state-of-the-art NVLINK interconnect technology, and a full pre-installed suite of the leading deep learning software, for developers to get a jump-start on deep learning research with the best tools that money can buy.
- Supports 2x Intel Xeon Scalable family processors
- 8x A100 40 GB SXM4 GPUs (19.5 TFlops of single precision, 1555 GB/s of memory bandwidth, and 40 GB of memory per board)
- NVIDIA DIGITS software providing powerful design, training, and visualization of deep neural networks for image classification
- Pre-installed standard Ubuntu 18.04 w/ Deep Learning software stack
- Google TensorFlow software library
- Automatic software update tool included
- A turn-key server with up to 600GB/s NVLINK interconnect
NVIDIA SXM4 GPU Options
|Model||Standard Memory||Memory Bandwidth (GB/s)||CUDA Cores||Tensor Cores||Single Precision (TFLOPS)||Double Precision (TFLOPS)||Power (W)||Explore|
|A100 40 GB SXM4||40 GB HBM2e||1555||6912||432||19.5||9.7||400||--|
EMLI (Exxact Machine Learning Images)
*Additional NGC (NVIDIA GPU Cloud) containers can be added upon request.
Conda EMLISeparated Frameworks
Container EMLIFlexible. Reconfigurable.
DIY EMLISimple. Clean. Custom.
Who is it for?
For developers who want pre-installed deep learning frameworks and their dependencies in separate Python environments installed natively on the system.
For developers who want pre-installed frameworks utilizing the latest NGC containers, GPU drivers, and libraries in ready to deploy DL environments with the flexibility of containerization.
For experienced developers who want a minimalist install to set up their own private deep learning repositories or custom builds of deep learning frameworks.
|Microsoft Cognitive Toolkit||—||—|
|NVIDIA CUDA Toolkit|
|NVIDIA CUDA Dev Toolkit||—|
|Micro-K8s||Free upgrade available||Free upgrade available||Free upgrade available|
- DDR4 SDRAM
- DDR4 NVDIMM (Intel Optane DCPMM)
- RAID 0, 1, 5, 10
- Via AIOM network card
- 8x PCI-E 4.0 x16 slots (Low-Profile, From PCI-E)
- 2x PCI-E 4.0 x16 slots (Low-profile, From CPU)
- 2x PCI-E M.2 slots
- 6x 2.5 hot-swap drive bays
- Supports NVMe drives
- 1x RJ45 Dedicated IPMI Port