Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


Ace Computers Adds NVIDIA Tesla P100 to HPC Clusters

tesla-p100Today Ace Computers from Illinois announced that the company is integrating the new Nvidia Tesla P100 accelerators into select HPC clusters.

Nvidia has been a trusted partner for many years,” said Ace Computers CEO John Samborski. “We are especially enthusiastic about this P100 accelerator and the transformative effect it promises to have on the clusters we are building for enterprises, academic institutions and research facilities.”

Large-scale datacenters rely on a significant number of interconnected commodity compute nodes, which limits the performance needed to drive leading-edge scientific and technological applications. The Tesla P100 taps into an innovative GPU architecture to deliver the world’s fastest compute node; capable of performance that equals 100s of slower commodity compute nodes.

The Tesla P100 has four groundbreaking technologies:

  • New Pascal Architecture: Delivering 5.3 and 10.6 teraflops of double and single precision for HPC, 21.2 teraflops of FP16 for deep learning.
  • NVIDIA NVLink: The world’s first high-speed interconnect for multi-GPU scalability with 5x boost in performance.
  • CoWoS with HBM2: Unifying data and compute into a single package for up to 3x memory bandwidth over prior-generation solutions.
  • Page Migration Engine: Parallel programming becomes simpler by enabling datasets beyond the physical limits of GPU memory.

Tesla P100 Accelerator Specifications:

  • 5.3 teraflops double-precision performance with NVIDIA GPU Boost
  • 10.6 teraflops single-precision performance with NVIDIA GPU Boost
  • 21.2 teraflops half-precision performance with NVIDIA GPU Boost
  • 160 GB/sec bi-directional interconnect bandwidth with NVIDIA NVLink
  • 732 GB/sec memory bandwidth with CoWoS HBM2 stacked memory
  • 16 GB of CoWoS HBM2 stacked memory
  • Enhanced programmability with Page Migration Engine and unified memory
  • ECC protection for increased reliability
  • Server-optimized for best throughput in the datacenter

Primary uses for Ace HPC clusters include: analytics; structured and unstructured data analysis; simulation/modeling; prototype design; existing design optimization; algorithm development; productivity efficiency; data management; and digital graphics design and implementation.

Nvidia has a stellar reputation for delivering innovations that are truly revolutionary,” Samborski said. “We look forward to delivering leading-edge clusters, powered by these new GPUs, to our clients.”

Sign up for our insideHPC Newsletter

Resource Links: