A regression‐based performance prediction framework for synchronous iterative algorithms on general purpose graphical processing unit clusters. (5th April 2013)

• Record Type: Journal Article
• Authors: Pallipuram, Vivek K.; Smith, Melissa C.; Raut, Nimisha; Ren, Xiaoyu
• Is Part Of: Concurrency and computation. Volume 26:Number 2(2014:Feb.)
• Page Start: 532
• Page End: 560
• Publication Date: 2013-04-05
• Subjects: Parallel processing (Electronic computers) -- Periodicals; Parallel computers -- Periodicals; 004.35
• DOI: 10.1002/cpe.3017
• Languages: English
• ISSNs: 1532-0626

Abstract

<abstract abstract-type="main" id="cpe3017-abs-0001"> <title>SUMMARY</title> <p id="cpe3017-para-0001">Heterogeneous performance prediction models are valuable tools to accurately predict application runtime, allowing for efficient design space exploration and application mapping. The existing performance models require intricate system architecture knowledge, making the modeling task difficult. In this research, we propose a regression‐based performance prediction framework for general purpose graphical processing unit (GPGPU) clusters that statistically abstracts the system architecture characteristics, enabling performance prediction without detailed system architecture knowledge. The regression‐based framework targets deterministic <italic>synchronous iterative algorithms</italic> using our <italic>synchronous iterative GPGPU execution model</italic> and is broken into two components: the <italic>computation component</italic> that models the GPGPU device and host computations and the <italic>communication component</italic> that models the network‐level communications. The computation component regression models use algorithm characteristics such as the number of floating‐point operations and total bytes as predictor variables and are trained using several small, instrumented executions of synchronous iterative algorithms that include a range of floating‐point operations‐to‐byte requirements. The regression models for network‐level communications are developed using micro‐benchmarks and employ data transfer size and processor count as predictor variables. Our performance prediction framework achieves prediction accuracy over 90% compared with the actual implementations for several tested GPGPU cluster configurations. The end goal of this research is to offer the scientific computing community, an accurate and easy‐to‐use performance prediction framework that empowers users to optimally utilize the heterogeneous resources. Copyright © 2013 John Wiley & Sons, Ltd.</p> </abstract>