Twitter
Facebook
You Tube
Blog
Forum

Novatech Deep Learning workstations

Optimised for DL workloads

A range of powerful, cost-effective and highly flexible workstations, designed for development and testing, before full-scale training at a data centre level.

NVIDIA Preferred Solutions Provider

As an NVIDIA Preferred Solutions Provider, we work closely with NVIDIA to combine our expertise and services to design, supply, and implement advanced NVIDIA computing solutions.

TRUSTED BY

- Filtered by : "Deep Learning"

ProStation DL9-2R - Development Workstation

Based on the Intel Core X-series processor platform, and paired with dual NVIDIA RTX 2080 Ti's offering up to 26 TFLOPS of FP32 compute performance. The ProStation DL9-2R is optimised for deep-learning development, where models can be created and optimised at an affordable workstation level.

The ProSation DL9-2R is fully customisable, offering the flexibility to accommodate up to four double slot GPU's, while maintaining 16 PCI-E lanes to each GPU.

All of our systems are thoroughly tested for any potential thermal throttling and are available pre-installed with Ubuntu, and any framework you require, including CUDA, DIGITS, Caffe Pytorch, Tensorflow, Theano, and Torch.

  • Intel Core i9 9820X 3.3GHz (4.1GHz Turbo) 10 Core Processor
  • 2x Asus RTX 2080 Ti Turbo 11GB Graphics (4352 CUDA Cores per GPU)
  • 64GB ECC DDR4 2400MHz quad-channel Memory
  • 512GB NVMe M.2 SSD + 4TB Enterprise-class HDD
  • Support for up to 4 double-width GPUs @ PCIe x16 (PLX switched)
  • 2x 10 Gigabit Ethernet

Order code: PC-2422

ProStation DLE-2R - Development Workstation

Featuring the AMD Ryzen Threadripper processor, with 24 cores and 64 PCI-E lanes for increased bandwidth, plus dual NVIDIA RTX 2080 Ti's offering up to 26 TFLOPS of FP32 compute performance. The ProStation DLE-2R is optimised for deep-learning development, where models can be created and optimised at an affordable workstation level.

The ProStation is fully customisable, offering the flexibility to accommodate up to four double slot GPU's, while maintaining 16 PCI-E lanes to each GPU.

All of our systems are thoroughly tested for any potential thermal throttling and are available pre-installed with Ubuntu, and any framework you require, including CUDA, DIGITS, Caffe Pytorch, Tensorflow, Theano, and Torch.

  • AMD Threadripper 2970WX 3GHz (4.2GHz Boost) 24-Core Processor
  • 2x Asus RTX 2080 Ti Turbo Graphics (4352 CUDA cores per GPU)
  • 64GB ECC DDR4 2400MHz quad-channel Memory
  • 512GB NVMe M.2 SSD + 4TB Enterprise-class HDD
  • Support for up to 4 double-width GPUs @ PCIe x16

Order code: PC-2421

HyperStation DLE-3R - I/O throughput optimised GPU workstation

Featuring a single, 32-core AMD EPYC processor, the HyperStation DLE-3R is a high-performance, cost-effective deep learning solution in comparison to dual-processor platforms.

With the ability to address up to 128 PCIe lanes and 8-channel memory, the AMD EPYC platform offers superior memory, and I/O throughput allowing for flexibility in extending storage, networking or compute to a level previously unattainable.

Our default specification with 3x RTX 2080 Ti GPU's can deliver up to 40 TFLOPS of FP32 compute performance, with the ability to scale up to a maximum of 4 GPU's, to deliver accelerated AI, Deep Learning, Machine Learning, research and analytics workloads.

All of our systems are thoroughly tested for any potential thermal throttling and are available pre-installed with Ubuntu, and any framework you require, including CUDA, DIGITS, Caffe Pytorch, Tensorflow, Theano, and Torch.

  • AMD EPYC 7452 3GHz (4.2GHz Boost) 24-Core Processor
  • 3x Asus RTX 2080 Ti Turbo Graphics (4352 CUDA cores per GPU)
  • 128GB ECC DDR4 2666MHz 8-channel Memory
  • 512GB NVMe M.2 SSD + 2x 4TB Enterprise-class HDD
  • Support for up to 4 double-width GPUs @ PCIe x16
  • 2x 10 Gigabit Ethernet

Order code: PC-2420

HyperStation DLX-4R -

Featuring a 14-core, high-performance Intel Xeon W processor, up to four GPUs, and 512GB DDR4-2666MHz, the HyperStation DLX-4R delivers the power to train and run models of a more complicated nature, using much larger data sets.

Intel Xeon Scalable processors take AI performance to the next level with Intel Deep Learning Boost (Intel DL Boost), a new set of embedded processor technologies designed to accelerate AI workloads such as image recognition, object detection, speech recognition and language translation. It extends Intel Advanced Vector Extensions 512 (Intel AVX-512) with a new Vector Neural Network Instruction (VNNI) that significantly increases deep learning inference performance over previous generations.

Our default specification with 4x RTX 2080 Ti GPU's can deliver up to 54 TFLOPS of FP32 compute performance. With the ability to scale up to a maximum of 4x Tesla V100 GPU's, for the most demanding development and training needs.

All of our systems are thoroughly tested for any potential thermal throttling and are available pre-installed with Ubuntu, and any framework you require, including CUDA, DIGITS, Caffe Pytorch, Tensorflow, Theano, and Torch.

  • Intel Xeon W-2175 2.5Hz (4.3GHz Turbo) 14 Core Processor
  • 4x NVIDIA RTX 2080 Ti Turbo 11GB Graphics (4352 CUDA Cores per GPU)
  • 128GB ECC DDR4 2666MHz quad-channel Memory
  • 512GB NVMe M.2 SSD + 2x 4TB Enterprise-class HDD
  • Support for up to 4 double-width GPUs @ PCIe x16 (PLX switched)
  • 2x 10 Gigabit Ethernet

Order code: PC-2419

HyperStation DLX-4R - Dual Scalable Xeon Development Platform

Built on Intel's flagship Scalable Xeon platform the HyperStation DLX-4R offers not only outstanding performance and reliabiliy but also the flexibility to accommodate up to four double slot GPU's maintaining 16 PCI-E lanes to each GPU. Designed to allow developers to train, infer and scale their software on a powerful, near-silent multi GPU desktop solution before taking the next step and investing in expensive rack hardware.

Fully configurable and powered with Nvidia's latest Turing architecture this system accelerates diverse workloads including high-performance computing, deep learning training and inference, machine learning, data analytics, and graphics.

  • 2x Intel Xeon Silver 4114 Processor
  • 192GB DDR4 2400Mhz Samsung Registered DIMM Server Memory
  • 4 x NVIDIA Tesla T4 16GB Turing Advanced Inference Accelerator Card
  • Samsung 512GB PM981 M.2 PCIe NVMe M.2 SSD
  • Silent Design

Order code: PC-2418

HyperStation DLX-2V - Dual Xeon Gold Scalable platform

The HyperStation DLX-2V is a powerful, enterprise-class workstation, featuring Intel Dual Xeon Gold processors, and dual NVIDIA Quadro GV100 graphics to meet the demands of AI, photorealistic rendering, simulation, and VR workflows.

The base configuration utilises NVLink, to achieve up to 200GB/s direct GPU-to-GPU interconnect (up to 46% faster vs PCIe), offering a combined 64GB of HBM2 VRAM, and up to 30 TFLOPS of compute performance at FP32 and 118.5 TFLOPs of tensor operation capability, to support a wide range of compute-intensive workloads flawlessly.

Fully customisable and scalable, the HyperStation DLX-2V has the flexibility to accommodate up to four double slot GPU's in four-way NVLink architecture; allowing developers to train, infer and scale their software on a desktop solution before taking the next step and investing in expensive rack hardware.

All of our systems are thoroughly tested for any potential thermal throttling and are available pre-installed with Ubuntu, and any framework you require, including CUDA, DIGITS, Caffe Pytorch, Tensorflow, Theano, and Torch.

  • 2x Intel Xeon Gold 5120 2.2Hz (3.2GHz Turbo) 14 Core per Processor
  • 2x NVIDIA Quadro GV100 32GB HMB2 Graphics (5120 CUDA Cores per GPU)
  • 192GB ECC DDR4 2400MHz quad-channel Memory
  • 512GB NVMe M.2 SSD + 2x 4TB Enterprise-class HDD
  • Support for up to 4 double-width GPUs @ PCIe x16
  • 2x 10 Gigabit Ethernet

Order code: PC-2417

Nvidia DGXS-1 includes 1st year SUMS Support Package

  • Intel® Xeon® Processor E5-2698 v4
  • 256GB DDR4 RDIMM Memory
  • 4 x 1.92 TB SSD RAID 0
  • 4 x NVIDIA Tesla V100 32GB GDDR5

Order code: SR-0260

We use cookies to provide a better customer experience on our site. By continuing to shop with us, you consent to our use of cookies. Our cookie policy is available here