Large System Performance of SPEC OMP2001 Benchmarks

Saito, Hideki; Gaertner, Greg; Jones, Wesley B.; Eigenmann, Rudolf; Iwashita, Hidetoshi; Lieberman, Ron; Waveren, G. Matthijs van; Whitney, Brian

doi:10.1007/3-540-47847-7_34

Cited by 25 publications

(19 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In their findings, all applications scale well with the exception of swim and apsi (which is not in line with our results, as well as, e.g. [7]). This study also evaluates "OpenMP overhead" by counting the number of parallel regions and multiplying this number with an empirically determined overhead for creating a parallel region derived from an execution of the EPCC micro-benchmarks [2].…”

Section: Related Workcontrasting

confidence: 64%

“…Due to space limitations we omit a textual description of the background, purpose, and implementation of each application, please refer to [7] for such a description. Instead, Table 1 lists the main characteristics of each application with respect to the OpenMP constructs used for parallelization (suffix m denotes the medium variant, while suffix l denotes the large variant of each application).…”

Section: The Spec Openmp Benchmarksmentioning

confidence: 99%

“…Saito et al [7] analyze the published results of the SPEC-OMPM suite on large machines (32 processors and above) and describe planned changes for the -then upcoming -large variant of the benchmark suite. A paper of Sueyasu [8] analyzes the scalability of selected components of SPEC-OMPL in comparison with the medium variant.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Scalability Analysis of the SPEC OpenMP Benchmarks on Large-Scale Shared Memory Multiprocessors

Fürlinger

Gerndt

Dongarra

2007

Computational Science – ICCS 2007

View full text Add to dashboard Cite

show abstract

Section: Related Workcontrasting

confidence: 64%

Section: The Spec Openmp Benchmarksmentioning

confidence: 99%

See 1 more Smart Citation

Scalability Analysis of the SPEC OpenMP Benchmarks on Large-Scale Shared Memory Multiprocessors

Fürlinger

Gerndt

Dongarra

2007

Computational Science – ICCS 2007

View full text Add to dashboard Cite

show abstract

“…The second, large suite, SPEC OMPL2001, focusing on 32-way and larger systems, was released in May 2002. SPEC OMPL2001 shares most of the application code base with SPEC OMPM2001, but the code and the data sets have been improved and made larger to achieve better scaling and also to reflect the class of computation regularly performed on such large systems [6]. So far, SPEC OMP performance has been reported for systems up to 128 processors.…”

Section: Introduction Spec (The Standard Performance Evaluation Comentioning

confidence: 99%

SPEC HPG benchmarks for high-performance systems

Müller

Kalyanasundaram

Gaertner³

et al. 2004

IJHPCN

Self Cite

View full text Add to dashboard Cite

Abstract-In this paper, we discuss results and characteristics of the benchmark suites maintained by the Standard Performance Evaluation Corporation's (SPEC) High-Performance Group (HPG). Currently, SPEC HPG has two lines of benchmark suites for measuring performance of large-scale systems: SPEC OMP and SPEC HPC2002. SPEC OMP uses the OpenMP API and includes benchmark suites intended for measuring performance of modern shared memory parallel systems. SPEC HPC2002 uses both OpenMP and MPI, and thus it is suitable for distributed memory systems, shared memory systems, and hybrid systems. SPEC HPC2002 contains benchmarks from three popular application areas, Chemistry, Seismic, and Weather Forecasting. Each of the three benchmarks in HPC2002 has a small and a medium data set, in order to satisfy the need for benchmarking a wide range of high-performance systems. We analyze published results of these benchmark suites regarding scalability. We also present current efforts of SPEC HPG to create new releases of the benchmark suites.

show abstract

“…Thus, multiple trace files are generated for each PinPoint. We have successfully traced fourthreaded runs of some SPECOMP2001 [10] programs this way.…”

Section: Simulating Multi-threaded Programsmentioning

confidence: 99%