Penguin Computing Announces New High-Density Computing Solutions

Powerful new lineup of blade servers and a dual-node, 1U rack mount server offer industry leading density and price / performance

  • Share on TwitterShare on FacebookShare on Google+Share on LinkedInEmail a friend
To support the latest data center designs and high performance computing applications, today's servers should meet three criteria: fast, dense, and economical. These exciting new offerings from Penguin more than fit such descriptors and are being offered to meet the demand for more compute power in less space

San Francisco, CA (PRWEB) October 29, 2008

Penguin Computing, specialists in application-optimized High Performance Computing (HPC) solutions, today announced the immediate availability of new server products that provide users industry leading per-rack density and performance utilizing the latest AMD and Intel processors. These new servers can be purchased individually or as compute nodes in one of Penguin Computing's Emperor Series Linux clusters.

"To support the latest data center designs and high performance computing applications, today's servers should meet three criteria: fast, dense, and economical. These exciting new offerings from Penguin more than fit such descriptors and are being offered to meet the demand for more compute power in less space", says Penguin Computing president and CEO, Charles Wuischpard.

BladeRunner II is Penguin's new blade server platform which combines high physical density with cutting-edge RAS (Reliability, Availability, and Serviceability) features. This system can provide up to 960 cores and 3.84TB of memory per rack. BladeRunner II is well suited for HPC deployments with high-availability requirements, as well as large, mission critical enterprise applications. This platform also provides optional, integrated, high performance networking options: either 10 Gigabit Ethernet or Double Data Rate (DDR) InfiniBand. These are the same blade servers currently powering the famous CERN Large Hadron Collider Project in Geneva, Switzerland.

The Intel Xeon blade server provides two Xeon 5400 sockets with up to 64GB RAM and six hot-pluggable SAS/SATA 2.5" enterprise grade drives. Unlike most blade architectures, this blade server is based on the latest Intel 5400 'Seaburg' chipset and supports advanced performance features such as a 1600MHz front side bus and Fully Buffered DDR2-800 RAM.

The AMD blade server doubles the number of CPU cores per rack unit by supporting up to four CPU sockets. By utilizing AMD Opteron 8300 series quad core CPUs, this configuration provides more CPUs per rack than any other conventional form factor. This is an ideal choice for cycle hungry, scalable HPC applications, which need maximum horsepower in a constrained physical environment.

The Relion 1672 is Penguin's new high-density Intel-based, 1U, rack mountable platform designed to provide the most economical computing power from the least amount of rack space. Each R1672 system includes two complete, independent nodes based on dual-socket Intel 5400 Series architecture effectively doubling the number of CPUs, available memory bandwidth and external I/O bandwidth available in 1U form factor. Both nodes can be populated with up to 64GB of RAM (128GB total) and the optional FB DDR2-800 RAM boosts memory access performance. This new platform is designed to deliver maximum distributed application performance in limited physical space for those desiring rack-mounted servers.

"These systems are already in use with a number of our customers and we believe are superior to all other high-density solutions in the number of configurable features, the support for the latest processor, memory, disk drive and networking architectures, and the overall cost of ownership per TFLOP of performance. We've listened closely to our clients, heard their needs, and are excited to be able to provide them with such industry-leading performance and density", says Jussi Kukkonen, Penguin Computing's director, server product management.

About Penguin Computing
Penguin Computing, headquartered in San Francisco, California, specializes in complete, integrated High Performance Computing solutions. Penguin has been a successful innovator for over a decade, providing Linux solutions to a variety of industries. Penguin's staff, including the originator of the Beowulf Cluster architecture, has unsurpassed experience in delivering a powerful combination of fully integrated clusters, comprehensive cluster management software, and services. For more information about Penguin Computing and Penguin products, or to download Scyld ClusterWare 4.2 for a free 45-day evaluation, please go to http://www.penguincomputing.com .

Penguin Computing is a registered trademark, and Scyld ClusterWare is a trademark of Penguin Computing, Inc. Linux is a registered trademark of Linus Torvalds. Other names are for informational purposes only and may be trademarks of their respective owners.

###


Contact