Penguin Computing Showcases OCP Platforms for HPC at SC15

Print Friendly, PDF & Email

Phil Pokorny, CTO, Penguin Computing

Phil Pokorny, CTO, Penguin Computing

In this video from SC15, Phil Pokorny from Penguin Computing describes the company’s Tundra HPC systems based on the Open Compute Project design standard. In a deployment called CTS-1, the NNSA will deploy Penguin Tundra OCP servers based on Intel processors at three national labs using liquid cooling from Asetek.

“CTS-1 shows how the Open Compute and Open Rack design elements can be applied to high-performance computing and deliver similar benefits as its original development for Internet companies,” said Philip Pokorny, Chief Technology Officer, Penguin Computing. “We continue to improve Tundra for both the public and private sectors with exciting new compute and storage models coming in the near future.”

Tundra ES delivers the advantages of Open Computing in a single, cost-optimized, high-performance architecture. Organizations can integrate a wide variety of compute, accelerator, storage, network, software and cooling architectures in a vanity-free rack and sled solution. This allows them to build optimized Intel CPU, Phi, ARM or NVIDIA systems with the latest Penguin, Intel or Mellanox high-speed network technology for maximum performance.

Penguin also displayed their new Magna OCP servers based on the OpenPOWER platform.

Penguin Magna 2001 is our first system based on OpenPOWER technology, providing an alternative to established platform architectures. Like all Penguin Computing systems, Magna 2001 runs Linux operating system and Linux application stacks. Penguin Magna 2001 is targeted for customers who are developing applications and evaluating performance on OpenPOWER architecture. Magna 2001 is also a capable virtualization host with support for up to 2TB of RAM and 96 virtual CPU cores. Popular open source cloud management frameworks like OpenStack are also compatible with OpenPOWER based systems.

See our complete coverage of SC15 * Sign up for our insideHPC Newsletter