Automatic parallelization of irregular applications

Gutiérrez, Eladio; Asenjo, Rafael; Plata, Óscar; Zapata, Emilio L.

doi:10.1016/s0167-8191(00)00052-1

Cited by 22 publications

(11 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These irregular problems arise especially in the scientific domain as most of simulation algorithms present data unpredictability and irregularity. Examples of such algorithms include sparse matrix computations, computational fluid dynamics, image processing, molecular dynamics simulations, galaxy simulations, climate modeling and optimization problems [26].…”

Section: Irregular Algorithmsmentioning

confidence: 99%

A pattern language for parallelizing irregular algorithms

Monteiro

2010

Proceedings of the 2010 Workshop on Parallel Programming Patterns

View full text Add to dashboard Cite

Section: Irregular Algorithmsmentioning

confidence: 99%

A pattern language for parallelizing irregular algorithms

Monteiro

2010

Proceedings of the 2010 Workshop on Parallel Programming Patterns

View full text Add to dashboard Cite

“…The difficulties which have prevented the development of such techniques lie mainly in the huge complexity associated with pointers and dynamic data structures. In this sense, for Fortran77 codes -in which these data structures are not allowed -there are a large number of optimizing and parallelizing compilers which successfully deal with complex code transformation even in the presence of irregular array access [3]. 0190-3918/01$10.00 0 2001 IEEE However, new solutions are a must due to the wide acceptance of these programming languages and due to the fact that dynamic data structures are important tools to achieve good performance and simplify the development of complex codes.…”

Section: Motivationmentioning

confidence: 99%

Progressive shape analysis for real C codes

Corbera

Asenjo

Zapata

2001

International Conference on Parallel Processing, 2001.

Self Cite

View full text Add to dashboard Cite

Dynamic and pointer-based data structures are widely used in symbolic or irregular C codes. Howevel; there is still a lack of compiler techniques to deal with the automatic optimization of such codes. In this paper we take a first step towards this final objective: the automatic identification of the data structure used in the code. More precisely, we describe the framework and the compiler we have implemented to capture complex data structures generated, traversed, and modified in C codes. Our method assigns a Reduced Set of Reference Shape Graphs (RSRSG) to each sentence to approximate the shape of the data structure after the execution of such a sentence. With the properties and operations that define the behavior of our RSRSG, the method can accurately detect complex recursive data structures. The compiler makes a progressive analysis in which the level of detail is increased during the analysis when needed. Several experiments are carried out with complex data structures to validate the capabilities of our compiler.

show abstract

“…Two important lines of this research field are language and compiler technologies [3,18,19,24,27,30,31]. Nowadays, most parallelizing compilers are able to generate efficient parallel codes for regular applications.…”

Section: Introductionmentioning

confidence: 99%

Approaches Based on Permutations for Partitioning Sparse Matrices on Multiprocessors

Garz

Garcı́a

2005

J Supercomput

View full text Add to dashboard Cite

This paper introduces new approaches to the data distribution-partition problem for sparse matrices in a multiprocessor environment. In this work, the data partition problem of a sparse matrix is modeled as a Min-Max Problem subject to the uniformity constrain when the goal is to balance the load for both sparse and dense operations. This problem is NP-Complete and two heuristic solutions (ABO and GPB) are proposed. The key of ABO and GPB is to determine the permutation of rows/columns of the input sparse matrix to obtain a sorted matrix with a homogeneous density of nonzero elements. Due to the heuristic nature of the proposed methods their validation is carried out by a comparative study of the parallel efficiency of two types of problems (sparse and mixed) when ABO, GPB, Block, Cyclic and MRD data distributions are applied.

show abstract

Automatic parallelization of irregular applications

Cited by 22 publications

References 20 publications

A pattern language for parallelizing irregular algorithms

A pattern language for parallelizing irregular algorithms

Progressive shape analysis for real C codes

Approaches Based on Permutations for Partitioning Sparse Matrices on Multiprocessors

Contact Info

Product

Resources

About