GPU DEDICATED SERVERS

Compute. Transcode. Render. Mine.

Perfect for streaming to YouTube.

GPU Configurations starting at $59/mo.

Yes, Get my Server

Get instant access to computing power, graphics rendering, video transcoding, desktop virtualization and crypto mining.

data-center-gpu
caffe-logo
 data-center-gpu-ready
simulia-abaqus
mxnet_logo_2
gromacs
tensorflow-white-2
torch-logo-light
data-center-gpu-ready-app
 data-center-gpu-ready-app-namd-logo
ENTERPRISE GRADE GPU SERVERS
HP ENTERPRISE SERVERS
HP ENTERPRISE SERVERS

Your GPU configuration is installed on Hewlett Packard Enterprise servers, stress tested for 100% compatibility and stability.

CHOOSE YOUR DATA CENTER
CHOOSE YOUR DATA CENTER

Get a GPU dedicated server, deployed in one of our New York or Bucharest data centers.

LOW LATENCY NETWORK
LOW LATENCY NETWORK

Your server is connected to a custom built, low latency global network.

SUPPORT
SUPPORT

Get access to instant support, from real humans, available around the clock via phone or live chat.

UP TO 4 MATCHING GPU'S PER SERVER

Nvidia Geforce 3090

NVIDIA RTX 3070 / 3080 / 3090


NVIDIA’S GeForce RTX 30 Series graphics cards run on Ampere architecture, 2nd generation RTX, featuring several new technologies, from faster Ray Tracing and Tensor Cores to advanced streaming multiprocessors.

The GeForce RTX 30 Series GPUs are defined by their innovative thermal design that delivers almost 2x the cooling performance of the previous generation.

The world’s fastest graphics memory, GDDR6X, delivers remarkable performance that makes it perfect for resource-intensive applications such as AI, visualization, and gaming.


  • RTX 3070 Specifications

  • 8 GB GDDR6
  • 5888 CUDA Cores
  • 512 GB/s Max Bandwidth
  • NVIDIA GPU Boost

  • RTX 3080 Specifications

  • 10 GB GDDR6X
  • 8704 CUDA Cores
  • 760 GB/s Max Bandwidth
  • NVIDIA GPU Boost

  • RTX 3090 Specifications

  • 24GB GDDR6X
  • 10496 CUDA Cores
  • 936 GB/s Max Bandwidth
  • NVIDIA GPU Boost

Compatible with Linux, CUDA/OpenCL, KVM, Windows.


NVIDIA QUADRO RTX 5000 / 6000 / 8000


The NVIDIA Quadro RTX series gives you access to the well-known Turing™ chip architecture that reformed the work of millions of designers and creators.

Hardware-accelerated ray tracing, state-of-the-art shading, new AI-based abilities, all for enabling artists to increase their rendering capabilities.

The Turing Streaming Multiprocessor architecture features 4608 CUDA® cores, and together with the Samsung 24 GB GDDR6 memory, supports complex designs, 8K video content, and enormous architectural datasets.


  • QUADRO RTX 5000 Specifications

  • 16 GB GDDR6
  • 3072 CUDA Cores
  • 448 GB/s Max Bandwidth
  • NVIDIA GPU Boost

  • QUADRO RTX 6000 Specifications

  • 24 GB GDDR6
  • 4608 CUDA Cores
  • 672 GB/s Max Bandwidth
  • NVIDIA GPU Boost

  • QUADRO RTX 8000 Specifications

  • 48 GB GDDR6
  • 4608 CUDA Cores
  • 672 GB/s Max Bandwidth
  • NVIDIA GPU Boost

Compatible with Linux, CUDA/OpenCL, KVM, Windows.


Nvidia-Rtx-8000

QUADRO RTX 4000


Get access to the best performance and features from a single PCI-e slot with NVIDIA’S QUADRO RTX 4000.

State-of-the-art display and memory technologies combined with the Turing™ chip architecture delivers photorealistic single ray-traced rendering in a fraction of a second.

This GPU features RT Cores, optimized for ray tracing, and Tensor Cores, perfect for deep learning projects.

Now you can create authentic VR experiences and enjoy faster performance when it comes to your AI applications with a cost-effective solution.


  • QUADRO RTX 4000 Specifications

  • 8 GB GDDR6
  • 2304 CUDA Cores
  • 416 GB/s Max Bandwidth
  • NVIDIA GPU Boost

Compatible with Linux, CUDA/OpenCL, KVM, Windows.


NVIDIA A40 / A100


NVIDIA’s Ampere Architecture, the successor to Volta, and is the fundamental solution for AI acceleration, from the edge to the cloud.

The NVIDIA A40 chip enables multi-workload capabilities with ultra-modern features for ray-traced rendering, VR, and more. Second generation RT cores deliver 2X the throughput over the previous one, the third generation Tensor cores provide 5X more training capabilities, and the 48 GB GDDR6 memory is more than enough for engineers, data scientists, and their large datasets and workloads.

The NVIDIA A100 Tensor Core GPU is a revolutionary leap for AI, as it delivers unrivaled acceleration at every scale, with NVIDIA’s Multi-Instance GPU (MIG) technology that allows the efficient scaling of thousands of GPUs. The third generation Tensor cores provide up to 20X more performance, and the MIG technology lets multiple networks operate at the same time on a single A100 GPU, optimizing computing resources.


  • NVIDIA A40 Specifications

  • 48 GB GDDR6 with error-correcting code (ECC)
  • 10752 CUDA Cores
  • 696 GB/s Max Bandwidth
  • NVIDIA GPU Boost

  • NVIDIA A100 Specifications

  • 40 GB GDDR6
  • 4608 CUDA Cores
  • 1555 GB/s Max Bandwidth
  • NVIDIA GPU Boost
Nvidia-a100

NVIDIA TESLA T4


The T4 introduces Tensor Core technology with multi-precision computing, making it up to 40 times faster than a CPU and up to 3.5 times faster than its Pascal predecessor, the Tesla P4.


Get access to 8.1 TFLOPS of single precision performance from a single T4 GPU.


Transcode up to 38 full HD video streams simultaneously with a single Tesla T4 GPU paired with our HPE BL460c blade server.

*Results may vary, based on server configuration.

  • TURO TU104
  • 320 TURING TENSOR CORES
  • 2560 CUDA CORES
  • 16 GB GDDR6
  • 8.1 TFLOPS SINGLE PRECISSION
  • 65 FP16 TFLOPS
  • 130 INT8 TOPS
  • 260 INT4 TOPS
  • 320 GB/s Max Bandwidth

Compatible: VMWare ESXi, Citrix Xenserver, KVM, Linux, Windows.

The Coral USB Accelerator

You can now add an Edge TPU coprocessor to any Linux-based system with the Coral USB Accelerator designed by Google. The small ASIC chip provides high-performance ML inferencing with low power cost. For example, it can execute 100fps on MobileNet v2 models, while using very little power (500mA at 5V).


Specifications

  • ARM 32 Bit Cortex 32 MHz
  • Edge TPU ASIC (for Lite TensorFlow models)
  • USB 3.1 5Gb/s transfer speed

Compatible with Linux machines, Debian 6.0 or higher, or any derivative (such as Ubuntu 10.0+), but also with Raspberry Pi (213 Mode B/B+).

NVIDIA GeForce RTX 2080 / RTX 2080 Ti


NVIDIA’s new Turing chip architecture delivers up to six times the performance of previous generation GPU’s, with breakthrough technologies and next generation, ultra-fast GDDR6 memory.


RTX 2080 Specifications

  • 8 GB GDDR6
  • 2944 CUDA Cores
  • 448 GB/s Max Bandwidth
  • NVIDIA GPU Boost 4.0

RTX 2080 TI Specifications

  • 11 GB GDDR6
  • 2944 CUDA Cores
  • 616 GB/s Max Bandwidth
  • NVIDIA GPU Boost 4.0

Compatible with Linux, CUDA/OpenCL, KVM.

NVIDIA GeForce GTX 1080/1070 TI


NVIDIA’s previous chip architecture, great for mining, graphics rendering and computing. The NVIDIA Pascal architecture delivers excellent performance at a budget friendly price.


  • 8 GB DDR5
  • 2560 CUDA Cores
  • 320 GB/s Max Bandwidth
  • NVIDIA GPU Boost 3.0

Compatible with Linux, CUDA/OpenCL, KVM.

NVIDIA TESLA P4/P40/P100


A optimal chip for machine learning and video transcoding, can be found in the NVIDIA Tesla P4 and P100 GPU’s. NVIDIA’s Pascal chip architecture has been proven to be faster and more power efficient than its Maxwell predecessor.

Transcode up to 20 simultaneous video streams with a single Tesla P4 paired with our HPE BL460c blade server. *

A more powerful version of the Tesla P4 is the Tesla P40, with more than twice the processing power of the Tesla P4.

The Tesla P100 GPU, is most suitable for deep learning and remote graphics. With 18.7 TeraFLOPS of inference performance, a single Tesla P100 can replace over 25 CPU servers.


  • Pascal GP100 or GP104 chip
  • Up to 3584 CUDA cores
  • Up to 16 GB CoWoS
  • Enterprise grade hardware

Compatible: VMWare ESXi, Citrix Xenserver, KVM, Linux, Windows.

*Results may vary, based on server configuration and video resolution of each stream.

NVIDIA TITAN V


The first GPU to break the 100 teraflop barrier of deep learning performance. NVIDIA’s Volta chip, is up to 3x faster than it’s Pascal chip predecessor.

Your deep learning project design can now be a reality, with little investment. Get the maximum per machine deep learning performance, replacing up to 30 single CPU servers with just one Titan V configuration.

Use the Titan V for high performance computing, from predicting the weather to discovering or finding new energy sources. Get your results up to 1.5x faster than NVIDIA’s Pascal predecessor.


  • NVIDIA Volta Chip
  • 5120 CUDA cores
  • 640 Tensor Cores
  • 12 GB CoWoS Stacked HBM2
  • 653Gbps max bandwidth

Compatible: VMWare ESXi, Citrix Xenserver, KVM, Linux, Windows.

Why Primcast?

Add a GPU to HP enterprise hardware, designed specifically for use with GPU add-ons, eliminating incompatibility issues or poor/underperformance of hardware. Your services are deployed on our global low latency network, backed by a 99.9% uptime SLA and supported by GPU server experts, around the clock.

Create your Primcast GPU dedicated server account today

Create your Primcast GPU dedicated server account today.