A genetic algorithms approach to modeling the performance of memory-bound computations

Tikir, Mustafa M.; Carrington, Laura; Strohmaier, Erich; Snavely, Allan

doi:10.1145/1362622.1362686

Cited by 57 publications

(36 citation statements)

References 33 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Williams et al [37] proposed the Roofline model about a theoretical model for analyzing upper bounds of performance with given computational bottlenecks and memory bottlenecks. Tikir et al [35] proposed to use genetic algorithms to predict achievable bandwidth from cache hit rates for memory-bound HPC applications. Duan et al [16] proposed to use a hybrid Bayesian-neural network to predict the execution time of scientific workflow in the Grid environment.…”

Section: Related Workmentioning

confidence: 99%

Performance Analysis Tool for HPC and Big Data Applications on Scientific Clusters

Yoo

Koo

Cao

et al. 2016

Conquering Big Data With High Performance Computing

View full text Add to dashboard Cite

Big data is prevalent in HPC computing. Many HPC projects rely on complex workflows to analyze terabytes or petabytes of data. These workflows often require running over thousands of CPU cores and performing simultaneous data accesses, data movements, and computation. It is challenging to analyze the performance involving terabytes or petabytes of workflow data or measurement data of the executions, from complex workflows over a large number of nodes and multiple parallel task executions. To help identify performance bottlenecks or debug the performance issues in large-scale scientific applications and scientific clusters, we have developed a performance analysis framework, using state-ofthe-art open-source big data processing tools. Our tool can ingest system logs and application performance measurements to extract key performance features, and apply the most sophisticated statistical tools and data mining methods on the performance data. It utilizes an efficient data processing engine to allow users to interactively analyze a large amount of different types of logs and measurements. To illustrate the functionality of the big data analysis framework, we conduct case studies on the workflows from an astronomy project known as the Palomar Transient Factory (PTF) and the job logs from the genome analysis scientific cluster. Our study processed many terabytes of system logs and application performance measurements collected on the HPC systems at NERSC. The implementation of our tool is generic enough to be used for analyzing the performance of other HPC systems and Big Data workows.

show abstract

Section: Related Workmentioning

confidence: 99%

Performance Analysis Tool for HPC and Big Data Applications on Scientific Clusters

Yoo

Koo

Cao

et al. 2016

Conquering Big Data With High Performance Computing

View full text Add to dashboard Cite

show abstract

“…In order to carefully investigate performance, we started with a very simple memory intensive kernel based on [14]. Essentially, this benchmark measures the time needed to access data by looping over an array of a fixed size using a fixed stride.…”

Section: A Importance Of Environment Parameters and Code Optimizationsmentioning

confidence: 99%

Performance Analysis of HPC Applications on Low-Power Embedded Platforms

Stanisic

Videau

Cronsioe

et al. 2013

Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE), 2013

View full text Add to dashboard Cite

Abstract-This paper presents performance evaluation and analysis of well-known HPC applications and benchmarks running on low-power embedded platforms. The performance to power consumption ratios are compared to classical x86 systems. Scalability studies have been conducted on the Mont-Blanc Tibidabo cluster. We have also investigated optimization opportunities and pitfalls induced by the use of these new platforms, and proposed optimization strategies based on auto-tuning.

show abstract

“…The models have been used to provide accuracy of within 15% absolute error for prediction of production codes on real-world HPC applications [10]. For a more complete description of the other pieces of the framework, please see Carrington et al [12] and Tikir et al [13].…”

Section: B Profiling Identified Idiomsmentioning

confidence: 99%

Modeling and Predicting Performance of High Performance Computing Applications on Hardware Accelerators

Meswani

Carrington

Unat

et al. 2012

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

Self Cite

View full text Add to dashboard Cite

Abstract-Computers with hardware accelerators, also referred to as hybrid-core systems, speedup applications by offloading certain compute operations that can run faster on accelerators. Thus, it is not surprising that many of top500 supercomputers use accelerators. However, in addition to procurement cost, significant programming and porting effort is required to realize the potential benefit of such accelerators. Hence, before building such a system it is prudent to answer the question 'what is the projected performance benefit from accelerators for the workloads of interest?' We address this question by way of a performance-modeling framework that predicts realizable application performance on accelerators rapidly and accurately without going to the considerable effort of porting and tuning.The modeling framework first automatically identifies commonly found compute patterns in scientific applications which we term idioms, which may benefit by accelerator technology. Next the framework models the predicted speedup of those idioms if they were to be ported to and run on hardware accelerators. As a proof of concept we characterize two kinds of accelerators 1) the FPGA accelerators on a Convey HC-1 system and 2) an NVIDIA FERMI GPU accelerator. We model performance of the idioms gather/scatter and stream and our predictions show that where these occur in two full-scale HPC applications, Milc and HYCOM, gather/scatter speeds up by as much as 15X, and stream by as much as 14X, whereas the overall compute time of Milc improves by 3.4% and HYCOM by 20%. The cost of migrating data to/from the accelerator device can dwarf the benefit of speedup and hence we present models of data migration costs and its impact on the performance of Milc and HYCOM.

show abstract

A genetic algorithms approach to modeling the performance of memory-bound computations

Cited by 57 publications

References 33 publications

Performance Analysis Tool for HPC and Big Data Applications on Scientific Clusters

Performance Analysis Tool for HPC and Big Data Applications on Scientific Clusters

Performance Analysis of HPC Applications on Low-Power Embedded Platforms

Modeling and Predicting Performance of High Performance Computing Applications on Hardware Accelerators

Contact Info

Product

Resources

About