Automatic program parallelization

Banerjee, Utpal; Eigenmann, Rudolf; Nicolau, Alexandru; Padua, David

doi:10.1109/5.214548

Cited by 221 publications

(120 citation statements)

References 91 publications

Supporting

Mentioning

119

Contrasting

Order By: Relevance

“…Issues such as loop dependencies and pointer aliasing [5] need to be handled, and hence automatic tools are often semi-automatic, requiring extra help from the programmer. Probably the most successful semi-automatic parallelization paradigm is OpenMP [20].…”

Section: Automatic Parallelizationmentioning

confidence: 99%

Evaluating automatically parallelized versions of the support vector machine

Codreanu

Droge

Williams

et al. 2014

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYThe Support Vector Machine (SVM) is a supervised learning algorithm used for recognizing patterns in data. It is a very popular technique in Machine Learning and has been successfully used in applications such as image classification, protein classification, and handwriting recognition. However, the computational complexity of the kernelized version of the algorithm grows quadratically with the number of training examples. To tackle this high computational complexity we have developed a directive-based approach that converts a gradient-ascent based training algorithm for the CPU to an efficient GPU implementation. We compare our GPU-based SVM training algorithm to the standard LibSVM CPU implementation, a highly-optimized GPU-LIBSVM implementation, as well as to a directive-based OpenACC implementation. The results on different handwritten digit classification datasets demonstrate an important speed-up for the current approach when compared to the CPU and OpenACC versions. Furthermore, our solution is almost as fast and sometimes even faster than the highly optimized CUBLAS-based GPU-LIBSVM implementation, without sacrificing the algorithm's accuracy.

show abstract

Section: Automatic Parallelizationmentioning

confidence: 99%

Evaluating automatically parallelized versions of the support vector machine

Codreanu

Droge

Williams

et al. 2014

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…The direction ">" denotes a backward cross-iteration dependence. We refer to [2,4,22,21] for a thorough description of direction vectors. We make use of the following lemmas, given in these papers.…”

Section: Direction Vectorsmentioning

confidence: 99%

On the Interaction of Tiling and Automatic Parallelization

Pan

Armstrong

Bae

et al. 2008

OpenMP Shared Memory Parallel Programming

View full text Add to dashboard Cite

Abstract. Iteration space tiling is a well-explored programming and compiler technique to enhance program locality. Its performance benefit appears obvious, as the ratio of processor versus memory speed increases continuously. In an effort to include a tiling pass into an advanced parallelizing compiler, we have found that the interaction of tiling and parallelization raises unexplored issues. Applying existing, sequential tiling techniques, followed by parallelization, leads to performance degradation in many programs. Applying tiling after parallelization without considering parallel execution semantics may lead to incorrect programs. Doing so conservatively, also introduces overhead in some of the measured programs. In this paper, we present an algorithm that applies tiling in concert with parallelization. The algorithm avoids the above negative effects. Our paper also presents the first comprehensive evaluation of tiling techniques on compiler-parallelized programs. Our tiling algorithm improves the SPEC CPU95 floating-point programs by up to 21% over non-tiled versions (4.9% on average) and the SPEC CPU2000 Fortran 77 programs up to 49% (11% on average). Notably, in about half of the benchmarks, tiling does not have a significant effect.

show abstract

“…Parallelism [4][11] is one of the main sources for performance improvement in modern computing environments. Key techniques for automatic detection of parallelism are discussed in [5].…”

Section: Related Workmentioning

confidence: 99%

“…Parallelism is one of the main sources for performance improvement [4] [11] in modern computing environment. This is particularly true in the area of high performance computing, where the cost of parallelism (on thread creation, scheduling and communication) is usually negligible when compared to heavy workload.…”

Section: Introductionmentioning

confidence: 99%

A Cost-Aware Parallel Workload Allocation Approach Based on Machine Learning Techniques

Long

Fursin

Franke

2007

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Parallelism is one of the main sources for performance improvement in modern computing environment, but the efficient exploitation of the available parallelism depends on a number of parameters. Determining the optimum number of threads for a given data parallel loop, for example, is a difficult problem and dependent on the specific parallel platform. This paper presents a learning-based approach to parallel workload allocation in a costaware manner. This approach uses static program features to classify programs, before deciding the best workload allocation scheme based on its prior experience with similar programs. Experimental results on 12 Java benchmarks (76 test cases with different workloads in total) show that it can efficiently allocate the parallel workload among Java threads and achieve an efficiency of 86% on average.

show abstract

Automatic program parallelization

Cited by 221 publications

References 91 publications

Evaluating automatically parallelized versions of the support vector machine

Evaluating automatically parallelized versions of the support vector machine

On the Interaction of Tiling and Automatic Parallelization

A Cost-Aware Parallel Workload Allocation Approach Based on Machine Learning Techniques

Contact Info

Product

Resources

About