PN-logo@2x

Get 25% on servers with NVIDIA Tesla GPUs

Boost your infrastructure performance with NVIDIA Tesla Volta and Pascal architectures. Now available at special prices!

Access Massive GPU Processing Power Globally in the Cloud.

Graphics Processing Units (GPUs) are the front-runners to support today’s growing demand for infrastructure parallel processing abilities. From Artificial Intelligence (AI), Machine Learning, Deep Learning, Big Data manipulation, 3D rendering, and even streaming, the needs for high-performance GPUs is undeniable.

NVIDIA Tesla Volta and Pascal architectures are built to allow for seamless data processing and are ideal for Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) workloads.

Power Meets Performance

Build-your-own “DGX-1 like” systems with 4x or 8x NVIDIA Tesla V100 with NVLink interconnection getting access to a massive amount of deep learning compute power. With up to 128 GB HBM2, 40960 CUDA cores and a whopping 1 PFLOP=1000 TFLOPS (tensor performance – mixed precision FP16/FP32), or 125 TFLOPs (FP32 – single precision performance) of computational power supported by a reliable phoenixNAP 10Gbps network.

The available GPU-ready platforms include:

NVIDIA Tesla V100
(PCIe)

  • NVIDIA Volta GPU architecture
  • 14 TeraFLOPS single-precision performance
  • 112 TeraFLOPS Tensor performance (mixed precision)
  • 5120 CUDA cores
  • 640 Tensor cores
  • 16 GB HBM2 Memory
  • Memory bandwidth 900 GB/s
  • Interconnect bandwidth 32 GB/s (Bi-Directional)
  • PCIe 3.0 interface

NVIDIA Tesla V100
(PCIe)

  • NVIDIA Volta GPU architecture
  • 15.7 TeraFLOPS single-precision performance
  • 125 TeraFLOPS Tensor performance (mixed precision)
  • 5120 CUDA cores
  • 640 Tensor cores
  • 16 GB HBM2 Memory
  • Memory bandwidth 900 GB/s
  • Interconnect bandwidth 300 GB/s (Bi-Directional)
  • NVLink interface

NVIDIA Tesla V100
(PCIe)

  • NVIDIA Pascal GPU architecture
  • 12 TeraFLOPS single-precision performance
  • 3840 CUDA cores
  • 24GB DDR5 Memory
  • Memory bandwidth 346 GB/s
  • PCIe 3.0 interface

IaaS Platform Access

Gain access to phoenixNAP’s entire IaaS platform: dedicated servers, cloud solutions, security services, hybrid block storage and fully S3 compatible object storage to Disaster Recovery & Backup Services.

Dedicated GPUs

Don’t settle for anything less than fully dedicated GPUs physically attached to your servers. No virtualization; no shared resources. Full access to maximum NVIDIA Tesla GPU power and architectures.

No Upfront Investment

Access the most powerful GPU infrastructure with complete Capex avoidance, no upfront investments, and zero complications. Focus on developing your business and improving your bottom line.

© 2019 Copyright phoenixNAP | Global IT Services. All Rights Reserved. Privacy Policy