Penguin Computing's Scyld ClusterWare(TM) HPC Offers Enhanced High Availability and Most Advanced Linux Cluster Virtualization, Enabling Broader Cluster Use

BOSTON, LINUX WORLD, April 5 /PRNewswire/ -- Penguin Computing, the leader in Cluster Virtualization, today announced its new Scyld ClusterWare portfolio designed to drive productivity up and measurable cost out for commercial and public sector organizations running high performance computing (HPC) applications. Scyld ClusterWare, the latest evolution of the innovative architecture of Scyld Beowulf(R), provides a virtualized cluster environment that allows even non-system administrators to run a cluster. The new portfolio introduces support for multiple Linux platforms for broad compatibility with commercial applications and a high availability feature set to ensure maximum productivity in competitive HPC environments. The portfolio also includes the Life Sciences Suite comprising a central management 'dashboard' and the most frequently used life science applications ready to run, out of the box.

(Logo: http://www.newscom.com/cgi-bin/prnh/20051012/SFPENGCOMPLOGO )

"We make the complex simple, so users can focus on their work and not the technology. Scyld ClusterWare fundamentally changes the way large pools of servers are deployed and managed," said Enrico Pesatori, CEO of Penguin Computing. "This portfolio provides a hardened, commercial grade solution that gives commercial organizations and the public sector a predictable, simplified environment for running standard applications in the production technical arena."

Scyld's unique Cluster Virtualization architecture makes large pools of Linux servers appear and act like a single virtual system. Through a single point of command/control (the 'Master'), thousands of systems can be managed as if they were a single, consistent, virtual system, dramatically simplifying deployment and management and significantly improving server performance and data center resource utilization. With both commercial organizations and the public sector under ever increasing pressure to do more for less as budgets shrink and capacity requirements increase, Scyld ClusterWare's simplicity of management and scalability removes many of the ongoing costs associated with the complex management and administration of competitive offerings. Simple Deployment and Scalability On Demand

Scyld ClusterWare significantly improves system scalability while retaining overall simplicity of management. Through enhanced 'just-in-time' provisioning, where compute nodes only take from the Master the resources they require to run an application on an as needed basis, organizations can respond rapidly to changes in their business and provision and re-provision on the fly. With a Multi-Master architecture, Scyld ClusterWare's Multiple Masters can cooperate to enable expansion and contraction of a virtual pool of resources, on demand. Policy modules enable cooperative coordination and sharing of resources, ensuring optimal server utilization across the enterprise to deliver greater return on investment. Application Compatibility, Flexibility and Choice

Scyld ClusterWare has responded to the needs of HPC customers who need to deploy commercial independent software vendor applications in a predictable environment by offering support for multiple operating systems. Customers can now opt to run Scyld ClusterWare on top of either an integrated Linux operating system from CentOS or -- versions of Red Hat Enterprise Linux ES supported for use with Scyld ClusterWare. Support for standard Linux Operating systems allows customers to significantly increase the number of applications they can run with their Penguin cluster and choice enables easier integration into existing infrastructures. Enhanced High Availability

Scyld ClusterWare enables customers to maintain maximum productivity in competitive HPC environments. This is critically important since competitiveness is time-driven in HPC; time to discovery and time to production is dependent in part on the number of HPC 'runs' completed in that timeframe. With its new high availability feature set, Scyld ClusterWare ensures maximum uptime with seamless Master failover and maximum compute availability when compute nodes migrate between Masters, so jobs run uninterrupted and more jobs are completed within the allotted time. Life Sciences Suite

The new Life Sciences Suite comprises a central management 'dashboard' and the most frequently used life science applications pre-certified and ready to run, out of the box. These key open source bioinformatics applications can be managed within a web-based framework that provides a convenient environment for bio researchers to execute their analyses and manage data sets in an intuitive, secure and easy-to-use format. Examples of applications supported are NCBI BLAST, EMBOSS, Glimmer, ClustalW, HMMER, Wise, FastA, R, T-Coffee, MrBayes, Primer3 and MPIBLast. The Life Sciences Suite is built upon Terra Soft Solutions' Y-Bio v1.1, which offers an unprecedented level of efficiency, security, and organization for the bioinformaticist.

"Given the explosive growth of life sciences at an individual, lab, and institutional level, the timely success of a research project is now correlated to the efficiency of the suite of computational tools," said Kai Staats, CEO of Terra Soft Solutions. "Fortunately, Penguin's Life Sciences Suite using Y-Bio provides an unprecedented level of efficiency, security, and organization for the bioinformaticist. It removes the need for reliance upon HPC administrators, saving time, resources and money."

The innovative Scyld design features lightweight, diskless provisioning and the Scyld "single system" process space. In addition, Scyld ClusterWare provides support for integrated IPMI hardware management, MPICH , MVAPICH Silverstorm, and LAM MPI tools as well as PVFS and PVM file systems. An optimized Ganglia cluster monitoring system allows scalable system monitoring. Scyld ClusterWare also offers flexibility with GNU compilers included and support for PGI and Pathscale. Supported hardware platforms include Intel Pentium and Intel Xeon-based servers as well as AMD Opteron-based servers. About Penguin Computing

Penguin Computing is the leading provider of Cluster Virtualization solutions, delivering virtualized cluster systems to the High Performance Computing ("HPC") and Enterprise markets. Its virtualized cluster solutions are driven by Scyld ClusterWare(TM), its unique industry-leading Linux clustering software, which dramatically simplifies the deployment and management of Linux clusters by making large pools of servers act like a single, consistent, virtual system. As a result, Penguin Computing's clusters deliver the power, manageability and scalability of an expensive SMP machine or supercomputer, at a fraction of the cost. The company's extensive customer base includes Fortune 1000 companies, government agencies and educational institutions. Founded in 1998, Penguin Computing is headquartered in San Francisco, California. To learn more about HPC and Linux clustering, please visit Penguin Computing at http://www.penguincomputing.com . NOTE: Penguin Computing is a registered trademark and Scyld ClusterWare is a trademark of Penguin Computing, Inc. Scyld Beowulf is a registered trademark of Scyld Software Corporation. Linux is a registered trademark of Linus Torvalds. AMD, the AMD Arrow logo, AMD Opteron and combinations thereof, are trademarks of Advanced Micro Devices, Inc. Intel, Pentium and Xeon are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. Other names are for informational purposes only and may be trademarks of their respective owners.