Today at the International Supercomputing Conference, Microsoft debuted in the top 25 of the world’s top 500 largest supercomputers with the National Center for Supercomputing Applications (NCSA), which ranked at No. 23 with 68.5 teraflops.
The company also announced that the release candidate version of Windows HPC Server 2008 will be available for download in the last week of June.
Key features that enable Windows HPC Server 2008 to efficiently scale to thousands of cores include a new high-speed NetworkDirect RDMA, Microsoft’s new remote direct memory access interface, highly efficient and scalable cluster management tools, a service-oriented architecture (SOA) job scheduler, and cluster interoperability through standards such as the High Performance Computing Basic Profile (HPCBP) specification produced by the Open Grid Forum (OGF). At the show, Mellanox Technologies Inc. will demonstrate its new ConnectX InfiniBand cards achieving 2 µsec latency with 2 GB per second throughput using the new NetworkDirect interface.
The NCSA used the beta version of Windows HPC Server 2008 to achieve its 68.5 teraflops and 77.7 percent efficiency on 9,472 cores, making this facility one of the most powerful supercomputing systems in the world and the fastest Windows cluster to date.
Similarly, computer scientists at Umea University in northern Sweden, also working with the beta version of Windows HPC Server 2008 on their supercluster, achieved 46 teraflops and 85.5 percent efficiency on 5,376 cores, making their system the second-largest Windows cluster ever deployed and the fastest academic cluster in Sweden. Umea University will run the new supercomputer at its facility known as HPC2N. The university’s cluster employs 672 IBM blade servers, and also marks the first time that Windows HPC Server 2008 has been run publicly on IBM hardware.
Last fall Microsoft initiated a parallel computing initiative, a program creating a set of common development tools across multicore desktops and clusters with the goal of enabling parallelism for a broad set of commercial applications. In addition, the company and Intel Corporation recently announced a joint investment to create two new Universal Parallel Computing Research Centers at the University of California, Berkeley, and the University of Illinois at Urbana-Champaign, aimed at accelerating developments in mainstream parallel computing.