Penguin Computing

Penguin Computing is a relatively small company (100–200 employees) but one of the largest private suppliers of enterprise and high-performance computing and cloud computing solutions in North America and is based in Fremont, California. Penguin Computing pioneers the design, engineering, integration, and delivery of solutions that are based on open architectures and are made of non-proprietary components from a variety of OEM providers. The company's products range from high-end servers and workstations to networking technologies to digital storage to software solutions.

Contents

• Penguin computing showcases ocp platforms for hpc at sc15
• Penguin computing showcases ocp hardware for hpc
• References

Penguin Computing was an early contributor to the Open Compute Project (OCP), is a Platinum member of the project foundation, and one of a limited number of authorized solutions providers. Penguin Computing developed its Tundra™ Extreme Scale (Tundra ES) product line (announced, November 2015) to apply the benefits of OCP (including open technology, efficiency, and cost effectiveness) to high-performance computing. Penguin Computing also joined the OpenPOWER Foundation this year with its Magna line of servers built around the Power Architecture and featuring compliance with Open Compute Project Open Rack infrastructure, designed for virtualization workloads in the hyperscale data center.

In 2015, Penguin Computing was awarded a contract with the U.S. Department of Energy’s National Nuclear Security Administration (NNSA) tri-laboratory Commodity Technology Systems program, or CTS-1. Under the $39 million contract, Penguin Computing provides over 7 petaFLOPS of computing power at Los Alamos National Laboratory, Sandia National Laboratories, and Lawrence Livermore National Laboratory. Coincident with this contract win, Penguin Computing launched its Federal Division to focus specifically on this market space.

The CTS-1 contract also represents one of the first and largest deployments of Intel's Omni-Path high-performance communications architecture. In 2016, Penguin Computing's contributions were recognized with Intel's "Partner of the Year - HPC Technical Computing" award.

In 2003, Penguin Computing, at the time a Linux server company, acquired Scyld Software, a leader in Beowulf cluster management software. Among Penguin Computing's software solutions (offered under its Scyld™ brand) are:

The company also operates one of the first, successful, pay-as-you-go, HPC as a Service. The service, called Penguin Computing On-Demand (POD), allows individuals and organizations to utilize a HPC environment without having to invest in on-premise infrastructure while eliminating much of the performance, scalability, and security challenges associated with shared infrastructure and multi-tenant, cloud environments.

POD eliminates the virtualization overhead associated with other cloud computing environments in order to provide the best performance. This is achieved through bare-metal compute nodes, a high-speed and low-latency interconnect, and high-performance storage. POD has hundreds of ready-to-run, optimized applications, and POD offers HPC technical and workflow experts for support at no additional charge.

Penguin Computing emphasizes the end-to-end nature of its portfolio of offerings, which includes a wide range of Engineering Services.[5] These include site assessment and data center planning, custom cluster design based on applications and workflows, testing and benchmarking, on-premise deployment, third-party hardware and software integration, software application builds, systems administration, and Linux/HPC technical support.

Historical Timeline