Skip navigation

Standard Performance Evaluation Corporation

 
 

SPEC releases MPI benchmark suite based on actual end-user applications

WARRENTON, Va., July 17, 2007 — The Standard Performance Evaluation Corp. (SPEC) has released a benchmark suite that measures the performance of parallel computing systems and clusters running Message-Passing Interface (MPI) applications.

SPEC MPI2007 provides performance metrics that can be used to compare different hardware architectures (SMP, NUMA, clusters) and interconnects, processors, memory hierarchy, compilers, and MPI implementations. It tests performance based on actual end-user applications, instead of synthetic workloads or parallelized versions of sequential benchmarks.

“SPEC saw the need within the research and science communities for a standardized benchmark suite based on real MPI applications,” says Kalyan Kumaran, Ph.D., SPEC HPG chair and manager of applications performance engineering and data analytics at Argonne National Lab. “As with all of our benchmarks, SPEC MPI2007 benefits from SPEC’s nearly two decades of experience in establishing source code and data sets, measurement tools, run rules, and peer reviews that help ensure comparable and repeatable results.”

SPEC MPI2007 was developed by SPEC members AMD, Argonne National Lab, Fujitsu, Hewlett-Packard, IBM, Intel, QLogic, SGI, Sun, and the University of Dresden. It supports a wide range of 32- and 64-bit architectures, and is portable across Linux, UNIX and Windows operating systems. A medium-sized data set in SPEC MPI2007 scales up to 512 processors, and SPEC is working on larger data sets.

Two metrics are provided by SPEC MPI2007: A base metric for users who prefer the relative simplicity of a single-step build process, and a peak metric for those who want to experiment with different compilers and compiler flags to achieve the best performance. Performance measurement is normalized against a reference system to produce a “bigger-is-better” metric. The current reference system is an eight-node cluster of Celestica A2210 systems, each with dual AMD Opteron processors, connected by a TCP (GigE) interconnect.

Available immediately

SPEC MPI2007 is available immediately for $800 on DVD. It can be ordered online at the SPEC web site (www.spec.org) or by contacting the SPEC office by phone: 540-349-7878, fax: 540-349-5992, or e-mail: info@spec.org. More information and initial performance results for more than 20 systems running SPEC MPI2007 are available on the SPEC web site free of charge.

About SPEC

SPEC is a non-profit organization that establishes, maintains and endorses standardized benchmarks to measure the performance of the newest generation of high-performance computers. Its membership comprises leading computer hardware and software vendors, universities, and research organizations worldwide.

###
Media contact: Bob Cramblitt
Cramblitt & Company
919-481-4599; info@cramco.com

Product and service names mentioned herein may be the trademarks of their respective owners