Flexible workload generation for HPC cluster efficiency benchmarking

Molka, Daniel; Hackenberg, Daniel; Schöne, Robert; Minartz, Timo; Nagel, Wolfgang E.

doi:10.1007/s00450-011-0194-9

Cited by 19 publications

(10 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…loop unrolling) are adopted to improve performance/power relationship and energy efficiency for a given application as discussed in [17]. Previous efforts like [16], [17], [26] list some tools (like PowerTracer, SPEC PTDaemon and Intel Energy Checker SDK [10], [13], [15]) and best practices that could be used to measure power in high performance computing systems. An analysis of GPU GEM package using properties of CUDA on GPU to evaluate the GEM implementation using metrics such as performance and energy efficiency is discussed in [11].…”

Section: Related Workmentioning

confidence: 99%

Energy Analysis of Parallel Scientific Kernels on Multiple GPUs

Ghosh

Chandrasekaran

Chapman

2012

2012 Symposium on Application Accelerators in High Performance Computing

View full text Add to dashboard Cite

A dramatic improvement in energy efficiency is mandatory for sustainable supercomputing and has been identified as a major challenge. Affordable energy solution continues to be of great concern in the development of the next generation of supercomputers. Low power processors, dynamic control of processor frequency and heterogeneous systems are being proposed to mitigate energy costs. However, the entire software stack must be re-examined with respect to its ability to improve efficiency in terms of energy as well as performance.In order to address this need, a better understanding of the energy behavior of applications is essential. In this paper we explore the energy efficiency of some common kernels used in high performance computing on a multi-GPU platform, and compare our results with multicore CPUs. We implement these kernels using optimized libraries like FFTW, CUBLAS and MKL. Our experiments demonstrate a relationship between energy consumption and computation-communication factors of certain application kernels. In general, we observe that the correlation of energy consumption to GPU global memory accesses is 0.73 and power consumption to operations per unit time is 0.84, signifying a strong positive relationship between them. We believe that our results will assist the HPC community in understanding the power/energy behavior of scientific kernels on multi-GPU platforms.

show abstract

Section: Related Workmentioning

confidence: 99%

Energy Analysis of Parallel Scientific Kernels on Multiple GPUs

Ghosh

Chandrasekaran

Chapman

2012

2012 Symposium on Application Accelerators in High Performance Computing

View full text Add to dashboard Cite

show abstract

“…Some of the performance tracing tools already in existence, such as VampirTrace, are already extended to trace the power consumption using a wide variety of measurement devices. The visualization tool Vampir can display the traced-based information to analyse the power consumption [22] and the processor frequency [23], as well as other functions.…”

Section: Related Workmentioning

confidence: 99%

Tracing and Visualization of Energy-Related Metrics

Minartz

Kunkel

Ludwig

2012

2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops &Amp; PhD Forum

Self Cite

View full text Add to dashboard Cite

In an effort to reduce the energy consumption of high-performance computing centers, a number of new approaches have been developed in the last few years. One of these approaches is to switch hardware to lower power states in phases of device idleness or low utilization. Even if the concepts are already quite clear, tools to identify these phases in applications and to determine impact on performance and power consumption are still missing. In this paper, we investigate the tracing of energy-related metrics into our existing tracing environment in an effort to correlate them with the application. We implement tracing of performance and sleep states of the processor, the disk and the network device states in addition to the node power consumption. The exemplary energy efficiency analysis visually correlates the application with the energy-related metrics. With this correlation, it is possible to identify and further avoid waiting times caused by mode switches initiated by the user or the system.

show abstract

“…It is an academic attempt to evaluate the energy efficiency of a server across the workload space through synthetic workload generation. Molka et al [23] have a similar effort but for parallel workload generation. SERT [18] developed by SPEC is yet another example, and it targets server-class computer systems.…”

Section: Energy Benchmarksmentioning

confidence: 99%

Towards efficient supercomputing

Hsu

Kuehn

Poole

2012

Proceedings of the 3rd ACM/SPEC International Conference on Performance Engineering

View full text Add to dashboard Cite

Efficiency in supercomputing has traditionally focused on execution time. In early 2000's, the concept of total cost of ownership was re-introduced, with the introduction of efficiency measure to include aspects such as energy and space. Yet the supercomputing community has never agreed upon a metric that can cover these aspects completely and also provide a fair basis for comparison. This paper examines the metrics that have been proposed in the past decade, and proposes a vector-valued metric for efficient supercomputing. Using this metric, the paper presents a study of where the supercomputing industry has been and where it stands today with respect to efficient supercomputing.

show abstract

Flexible workload generation for HPC cluster efficiency benchmarking

Cited by 19 publications

References 11 publications

Energy Analysis of Parallel Scientific Kernels on Multiple GPUs

Energy Analysis of Parallel Scientific Kernels on Multiple GPUs

Tracing and Visualization of Energy-Related Metrics

Towards efficient supercomputing

Contact Info

Product

Resources

About