December 14, 2007
Marlborough, Mass., Dec. 6 -- Scali, the leader in higher performance computing software, today announced that the Standard Performance Evaluation Corp. (SPEC) has published new benchmark results provided by Scali using the SPEC MPI2007 benchmark suite on Microsoft Windows Server 2003 operating system. These are the first SPEC benchmarks released on a Windows-based platform that measure MPI-parallel, floating point, and compute intensive performance and demonstrate the viability of Windows-based solutions for high performance computing applications.
The benchmark was performed on a single node, Intel x86_64 based system. The quad-core Dell PowerEdge 1955 system was tested using Scali MPI Connect v5.5. Scali is the first vendor to release SPEC MPI2007 benchmarks for quad-core systems and the first to release both Linux and Windows based results. Scali also has the two highest per node(1) SPEC MPI2007 results on x86_64 based systems as of this release date.
"With the advent of multi-core systems, single workstations are quickly becoming a powerful and cost-effective alternative for running many HPC applications," said Håkon Bugge, CTO of Scali. "While Linux remains the clear leader in clustered computing, Windows is establishing itself not simply as a viable alternative for single-node deployments, but rather as the driving force in that segment."
Scali MPI Connect
Scali MPI Connect is a fully integrated message passing interface (MPI) solution, designed to enable applications to run at maximum performance through its unique, high performance MPI implementation. With Scali MPI Connect, application developers can take advantage of leading interconnect and multi-core technologies to build high performance clusters. Scali MPI Connect supports a wider range of customer environments while lowering the number of binaries required, and delivers superior application performance.
SPEC MPI2007 is SPEC's benchmark suite for evaluating MPI-parallel, floating point, and compute intensive performance across a wide range of cluster and SMP hardware. MPI2007 continues the SPEC tradition of giving HPC users the most objective and representative benchmark suite for measuring the performance of SMP (shared memory multi-processor) systems. SPEC reviews and publishes submitted results on their Web site at http://www.spec.org/mpi2007/results/.
Founded in 1997, Scali is the leading developer of higher performance clustering solutions that enable companies to maximize their investment in their performance datacenters. Hundreds of companies, including Goodyear, DaimlerChrysler, Lockheed Martin, US Army and Naval Air, Schlumberger and Shell, benefit from using Scali's cluster management solutions and MPI libraries. Scali helps these customers fully leverage their datacenter investments, respond quickly to evolving business requirements, lower IT costs and reduce datacenter complexity. The company is headquartered in Marlborough, Mass. with research and development in Oslo, Norway and Nagpur, India. Visit www.scali.com for more information.
(1) As of the date of this announcement, of all submitted SPECmpiM_base2007 results on x86_64 based clusters, the highest SPECmpiM_base2007 rates "per node" were obtained by Scali's two postings using quad-core systems. In order to establish a "per node rate" for the SPECmpiM_base2007 rate, Scali has compared all published results on x86_64 based clusters, and divided this particular rate by the number of nodes used to obtain the results. The derived metric, (i.e., SPECmpiM_base2007 per node) compares equal number of chips, while taking into account that the number of cores and MPI ranks on Scali's two quad-core postings using this metric are twice that of the other x86_64 based clusters.
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?