S12---The HPC Challenge (HPCC) benchmark suite

Łuszczek, Piotr; Bailey, David H.; Dongarra, Jack; Kepner, Jeremy; Lucas, Robert F.; Rabenseifner, Rolf; Takahashi, Daisuke

doi:10.1145/1188455.1188677

Cited by 273 publications

(121 citation statements)

References 0 publications

Supporting

Mentioning

118

Contrasting

Unclassified

Order By: Relevance

“…The system itself consists of 82,944 SPARC64 VIIIfx oct-core processors running at the clock speed of 2.0 GHz with 1.3 PB of memory. We measure the system's rack energy consumption for one hour under three different kinds of activities: when it is idle, when it runs HPL (High Performance Linpack) benchmark [22], and when it performs I/O transfer (checkpointing). The energy consumption measurement results are shown in Table 2.…”

Section: Energy Measurement Of a Large-scale Hpc Systemmentioning

confidence: 99%

Energy-Performance Modeling of Speculative Checkpointing for Exascale Systems

Amrizal

Uno²,

Sato

et al. 2017

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYCoordinated checkpointing is a widely-used checkpoint/restart protocol for fault-tolerance in large-scale HPC systems. However, this protocol will involve massive amounts of I/O concentration, resulting in considerably high checkpoint overhead and high energy consumption. This paper focuses on speculative checkpointing, a CPR mechanism that allows for temporal distribution of checkpointings to avoid I/O concentration. We propose execution time and energy models for speculative checkpointing, and investigate energy-performance characteristics when speculative checkpointing is adopted in exascale systems. Using these models, we study the benefit of speculative checkpointing over coordinated checkpointing under various realistic scenarios for exascale HPC systems. We show that, compared to coordinated checkpointing, speculative checkpointing can achieve up to a 11% energy reduction at the cost of a relatively-small increase in the execution time. In addition, a significant energy-performance trade-off is expected when the system scale exceeds 1.2 million nodes.

show abstract

Section: Energy Measurement Of a Large-scale Hpc Systemmentioning

confidence: 99%

Energy-Performance Modeling of Speculative Checkpointing for Exascale Systems

Amrizal

Uno²,

Sato

et al. 2017

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…Several h igh-profile efforts such as the Top500 List [15], the Green 500 List [16], the SPECPower bench mark [17], and power-performance evaluation of the HPCC benchmarks [18,19] have elevated the interest in energy efficiency for high-end systems and servers. Ge et al proposed the PowerPack [20] framework for measuring correlated power and performance data on large scale systems and we use this framework to collect the results presented.…”

Section: Energy Efficiency In Hpcmentioning

confidence: 99%

“…Us ing the Equations (13) and (18), we can formu late the Energy Efficiency Factor (EEF) more accurately, (19) Where ) Equation (19) contains two categories of parameters which d irectly impact perfo rmance and energy consumption: 1) mach ine dependent variables and 2) application dependent variables:…”

Section: E Energy Efficiency Factor (Eef)mentioning

confidence: 99%

See 1 more Smart Citation

Iso-Energy-Efficiency: An Approach to Power-Constrained Parallel Computation

Song

et al. 2011

2011 IEEE International Parallel &Amp; Distributed Processing Symposium

View full text Add to dashboard Cite

show abstract

“…We used the HPC Challenge (HPCC) benchmarks [6] to evaluate these environments. HPC Challenge examines performance of HPC architectures using kernels with memory access patterns more challenging than those of the High Performance Linpack (HPL) benchmark used in the TOP500 list.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of the HPC Challenge Benchmarks in Virtualized Environments

Łuszczek

Meek

Moore

et al. 2012

Euro-Par 2011: Parallel Processing Workshops

View full text Add to dashboard Cite

Abstract. This paper evaluates the performance of the HPC Challenge benchmarks in several virtual environments, including VMware, KVM and VirtualBox. The HPC Challenge benchmarks consist of a suite of tests that examine the performance of HPC architectures using kernels with memory access patterns more challenging than those of the High Performance Linpack (HPL) benchmark used in the TOP500 list. The tests include four local (matrix-matrix multiply, STREAM, RandomAccess and FFT) and four global (High Performance Linpack -HPL, parallel matrix transpose -PTRANS, RandomAccess and FFT) kernel benchmarks. The purpose of our experiments is to evaluate the overheads of the different virtual environments and investigate how different aspects of the system are affected by virtualization. We ran the benchmarks on an 8-core system with Core i7 processors using Open MPI. We did runs on the bare hardware and in each of the virtual environments for a range of problem sizes. As expected, the HPL results had some overhead in all the virtual environments, with the overhead becoming less significant with larger problem sizes. The RandomAccess results show drastically different behavior and we attempt to explain it with pertinent experiments. We show the cause of variability of performance results as well as major causes of measurement error.

show abstract

S12---The HPC Challenge (HPCC) benchmark suite

Cited by 273 publications

References 0 publications

Energy-Performance Modeling of Speculative Checkpointing for Exascale Systems

Energy-Performance Modeling of Speculative Checkpointing for Exascale Systems

Iso-Energy-Efficiency: An Approach to Power-Constrained Parallel Computation

Evaluation of the HPC Challenge Benchmarks in Virtualized Environments

Contact Info

Product

Resources

About