Request information     Related links     Evaluate Windows HPC Server 2008 View Windows HPC Server 2008 demos and videos     Proven Results National Supercomputing Center for Energy and the Environment (NSCEE) Sees 46 Percent Performance Gain with HPC System Cornell Uses High-Performance Computing to Speed Research, Extend Software Access University of Arizona Finds HPC Server 26% Faster than Linux, More Cost-Effective     Products Microsoft Windows HPC Server 2008 HPC Pack 2008 SDK

High-performance computing for academic research

Help your institution's scientists and researchers unlock new insights faster with a high-performance computing (HPC) platform that's familiar to users of Microsoft software and that doesn't require expensive hardware.

Power and agility for researchers

Research that requires computers to run heavy processing workloads—such as modeling, molecular simulation, advanced mathematics, and computational chemistry—can happen on Microsoft 64-bit Windows HPC Server 2008. In addition to getting the computing power you need to provide to researchers, you can build on the platform that offers a familiar interface and tools that help you customize, deploy, manage, and integrate with your existing IT infrastructure.

Key benefits

Built on the 64-bit Windows HPC Server 2008 platform, Microsoft high-performance computing helps to achieve:

• Faster research results. Increase your computing power and help remove roadblocks between you and the insights your research is targeting. With an architecture that directly bypasses operating system and TCP/IP overhead, NetworkDirect—the new Remote Direct Memory Access (RDMA) interface from Microsoft for high-speed, low-latency networks, such as those running on InfiniBand, 10-Gigabit Ethernet, and Myrine—achieves better performance through more efficient use of processor resources.
• Enhanced productivity. Accomplish more—in less time and with less effort—by using a platform that looks and works in a way that is familiar to people who use the Microsoft Office system and other Microsoft software.
• Performance that grows with your needs. When you build on the Windows HPC Server 2008 platform, you can efficiently scale to thousands of processing cores and take advantage of management tools that help systems administrators proactively monitor system health and maintain system stability.
• A flexible and future-ready platform. Reduce the complexity of integrating with existing clusters by using the HPC Pack Job Scheduler, which supports industry standards, such as Open Grid Forum (OGF) HPC Profile for job scheduler interoperability. The Subsystem for UNIX-based Applications (SUA) integrates applications to maximize existing investments while extending UNIX-based and LINUX-based applications to Windows-based systems.
• Fast deployment. Set up and configure your system using system templates that allow an HPC administrator to create standardized system images or apply patches that can be consistently deployed across the cluster.
• A system that runs smoothly. Use built-in diagnostic tools to quickly identify and diagnose hardware, software, or network problems across the cluster. The combination of Windows Failover Services and SQL Server database clustering provides head node redundancy in the event of a hardware failure. Additional capabilities, such as automatic handling of routine and redundant tasks, and intelligent reporting and monitoring are possible through the use of System Center Operations Manager 2007.

Customer success case studies

The High Performance Computing Center North (HPC2N) at Umeå University, Sweden, which has traditionally provided HPC support for Linux-based applications, deployed Windows HPC Server 2008 and achieved 46.04 teraflops with an efficiency rating of 85.6 percent. With this outstanding result, HPC2N looks forward to serving a broader range of customers.

To support a growing number of Windows-based parallel applications, the Center for Computing and Communication (CCC) at RWTH Aachen University, Aachen, Germany, upgraded from Windows Compute Cluster Server 2003 to Windows HPC Server 2008 and is taking advantage of increased network performance and other advanced features of the new product.

As a world leader in protein behavior research, the Daggett Research Group at the University of Washington (state) is always in need of more computing power. The group augmented its existing Linux-based high-performance computing resources with two new HPC systems based on Windows Compute Cluster Server 2003, which has enabled the group to make fundamental breakthroughs in simulation results analysis.