Parallelization Spectroscopy: Analysis of Thread-level Parallelism in HPC Programs

被引：1

作者：

Kejariwal, Arun ^{[1
]}

Cascaval, Calin ^{[1
]}

机构：

[1] Univ Calif Irvine, Irvine, CA 92717 USA

来源：

ACM SIGPLAN NOTICES | 2009年 / 44卷 / 04期

关键词：

Performance; Measurement; Parallelism; Multithreading; Loop Transofrmations;

D O I：

10.1145/1594835.1504221

中图分类号：

TP31 [计算机软件];

学科分类号：

081202 ; 0835 ;

摘要：

In this paper, we present a method - parallelization spectroscopy for analyzing the thread-level parallelism available in production High Performance Computing (HPC) codes. We survey a number of techniques that are commonly used for parallelization and classify all the loops in the case study presented using a sensitivity metric: how likely is a particular technique is successful in parallelizing the loop.

引用

页码：293 / 294

页数：2

共 9 条

[1]

Bikshandi G., 2006, Proceedings of the 2006 ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming PPoPP'06, P48, DOI 10.1145/1122971.1122981

[2]

Bose P, 2006, IEEE MICRO, V26, P5, DOI 10.1109/MM.2006.44

[3]

Kejariwal A., 2006, ICS, page, P24

[4] Optimistic Parallelism Requires Abstractions [J].

Kulkarni, Milind ;

Pingali, Keshav ;

Walter, Bruce ;

Ramanarayanan, Ganesh ;

Bala, Kavita ;

Chew, L. Paul .

PLDI'07: PROCEEDINGS OF THE 2007 ACM SIGPLAN CONFERENCE ON PROGRAMMING LANGUAGE DESIGN AND IMPLEMENTATION, 2007, :211-222

[5]

*LAM MPI, LAM MPI PAR COMP

[6] Modeling Optimistic Concurrency using Quantitative Dependence Analysis [J].

von Praun, Christoph ;

Bordawekar, Rajesh ;

Cascaval, Calin .

PPOPP'08: PROCEEDINGS OF THE 2008 ACM SIGPLAN SYMPOSIUM ON PRINCIPLES AND PRACTICE OF PARALLEL PROGRAMMING, 2008, :185-195

[7]

WOLFE MJ, 1989, P SUP 89 NOV

[8]

[No title captured]

[9]

[No title captured]

← 1 →