A methodology towards automatic performance analysis of parallel applications

Calzarossa, Maria Carla; Massari, Luisa; Tessera, Daniele

doi:10.1016/j.parco.2003.08.002

Cited by 7 publications

(11 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By clustering thread performance for different metrics, PerfExplorer should discover these relationships and which metrics best distinguish their differences. Calzarossa et al [15] proposes a top-down methodology towards automatic performance analysis of parallel applications: first, they focuses on the overall behavior of the application in terms of its activities, and then they consider individual code regions and activities performed within each code region. Calzarossa et al [15] utilizes clustering techniques to summarize and interpret the performance information by identifying patterns or groups of code regions characterized by a similar behavior.…”

Section: Related Workmentioning

confidence: 99%

“…Calzarossa et al [15] proposes a top-down methodology towards automatic performance analysis of parallel applications: first, they focuses on the overall behavior of the application in terms of its activities, and then they consider individual code regions and activities performed within each code region. Calzarossa et al [15] utilizes clustering techniques to summarize and interpret the performance information by identifying patterns or groups of code regions characterized by a similar behavior. Ahn et al [28] use several multivariate statistical analysis techniques to analyze parallel performance behavior, including cluster analysis and F-ratio, factor analysis, and principal component analysis.…”

Section: Related Workmentioning

confidence: 99%

“…Second, several previous efforts can only automatically identify critical bottlenecks with apriori knowledge specified in terms of either the execution patterns that represent situations of inefficient behaviors [2] [3] [4] [5] or the predefined performance hypotheses/thresholds [7] [9] or the decision tree classification trained by microbenchmarks [26]. Third, while a lots of existing tools [15] [19] [20] [21] [24] [26] [28] [35] [15] take exploratory or confirmatory data analysis approaches to automatically discovering relationships of relevant performance data, these efforts do not focus on locating performance bottleneck and uncovering their root causes of performance bottlenecks.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Automatic performance debugging of SPMD-style parallel programs

Liu

Zhan

et al. 2011

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Automatic performance debugging of parallel applications includes two main steps: locating performance bottlenecks and uncovering their root causes for performance optimization. Previous work fails to resolve this challenging issue in two ways: first, several previous efforts automate locating bottlenecks, but present results in a confined way that only identifies performance problems with apriori knowledge; second, several tools take exploratory or confirmatory data analysis to automatically discover relevant performance data relationships, but these efforts do not focus on locating performance bottlenecks or uncovering their root causes.The simple program and multiple data (SPMD) programming model is widely used for both high performance computing and Cloud computing. In this paper, we design and implement an innovative system, AutoAnalyzer, that automates the process of debugging performance problems of SPMD-style parallel programs, including data collection, performance behavior analysis, locating bottlenecks, and uncovering their root causes. AutoAnalyzer is unique in terms of two features: first, without any apriori knowledge, it automatically locates bottlenecks and uncovers their root causes for performance optimization; second, it is lightweight in terms of the size of performance data to be collected and analyzed. Our contributions are three-fold: first, we propose two effective clustering algorithms to investigate the existence of performance bottlenecks that cause process behavior dissimilarity or code region behavior disparity, respectively; meanwhile, we present two searching algorithms to locate bottlenecks; second, on a basis of the rough set theory, we propose an innovative approach to automatically uncovering root causes of bottlenecks; third, on the cluster systems with two different configurations, we use two production applications, written in Fortran 77, and one open source code-MPIBZIP2 (http://compression.ca/mpibzip2/), written in C++, to verify the effectiveness and correctness of our methods. For three applications, we also propose an experimental approach to investigating the effects of different metrics on locating bottlenecks.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic performance debugging of SPMD-style parallel programs

Liu

Zhan

et al. 2011

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

show abstract

“…Calzarossa et al [13] rank code regions based on their dispersion across the process space to identify the most promising optimization target. Phase profiling [14] can expose time-varying load distributions that would otherwise be hidden when performance metrics are summarized along the time axis.…”

Section: Introductionmentioning

confidence: 99%

Identifying the Root Causes of Wait States in Large-Scale Parallel Applications

Böhme

Geimer²,

Wolf

et al. 2010

2010 39th International Conference on Parallel Processing

View full text Add to dashboard Cite

Driven by growing application requirements and accelerated by current trends in microprocessor design, the number of processor cores on modern supercomputers is increasing from generation to generation. However, load or communication imbalance prevents many codes from taking advantage of the available parallelism, as delays of single processes may spread wait states across the entire machine. Moreover, when employing complex point-to-point communication patterns, wait states may propagate along far-reaching cause-effect chains that are hard to track manually and that complicate an assessment of the actual costs of an imbalance. Building on earlier work by Meira Jr. et al., we present a scalable approach that identifies program wait states and attributes their costs in terms of resource waste to their original cause. By replaying event traces in parallel both in forward and backward direction, we can identify the processes and call paths responsible for the most severe imbalances even for runs with tens of thousands of processes.

show abstract

“…In recent years, performance prediction for parallel architectures has attracted considerable attention, ranging from kernel benchmarking studies, application performance studies, performance modeling and detailed comparative analysis of newer architectures (e.g., [1][2][3][4][5][6][7][8]). Other related work includes automatic performance analysis [9,10] and automatic tuning [11]. Performance prediction across platforms is increasingly important for developers and scientists to determine the performance of their specific applications on a plethora of platforms in deciding the system that best fits their needs in the long run.…”

Section: Introductionmentioning

confidence: 99%

Parallel Performance Modeling using a Genetic Programming-based Error Correction Procedure

et al. 2007

View full text Add to dashboard Cite

Performance models of high performance computing (HPC) applications are important for several reasons. First, they provide insight to designers of HPC systems on the role of subsystems such as the processor or the network in determining application performance. Second, they allow HPC centers more accurately to target procurements to resource requirements. Third, they can be used to identify application performance bottlenecks and to provide insights about scalability issues. The suitability of a performance model, however, for a particular performance investigation is a function of both the accuracy and the cost of the model. A semi-empirical model previously published by the authors for an astrophysics application was shown to be inaccurate when predicting communication cost for large numbers of processors. It is hypothesized that this deficiency is due to the inability of the model adequately to capture communication contention (threshold effects) as well as other unmodeled components such as noise and I/O contention. In this paper we present a new approach to capture these unknown features to improve the predictive capabilities of the model. This approach uses a systematic model error-correction procedure that uses evolutionary algorithms to find an error correction term to augment the eXisting model. Four variations of this procedure were investigated and all were shown to produce better results than the original model. Successful cross-platform application of this approach showed that it adequately captures machine dependent characteristics. This approach was then successfully demonstrated for a second application, further showing its versatility.

show abstract

A methodology towards automatic performance analysis of parallel applications

Cited by 7 publications

References 13 publications

Automatic performance debugging of SPMD-style parallel programs

Automatic performance debugging of SPMD-style parallel programs

Identifying the Root Causes of Wait States in Large-Scale Parallel Applications

Parallel Performance Modeling using a Genetic Programming-based Error Correction Procedure

Contact Info

Product

Resources

About