Experience with efficient array data flow analysis for array privatization

Gu, Junjie; Li, Zhiyuan; Lee, Gyungho

doi:10.1145/263764.263785

Cited by 30 publications

(17 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(8) Panorama is a source-to-source translator for Fortran F77 programs, which implements automatic parallelization in addition to a number of loop transformation techniques, including loop fusion, loop shifting, loop tiling, and so on. We implemented a network flow algorithm, enhanced capacity scaling algorithm, (13) to solve Problem 4.…”

Section: Resultsmentioning

confidence: 99%

A Polynomial-Time Algorithm for Memory Space Reduction

Song

Wang

2005

Int J Parallel Prog

Self Cite

View full text Add to dashboard Cite

Reducing memory space requirement is important to many applications. For data-intensive applications, it may help avoid executing the program out-ofcore. For high-performance computing, memory space reduction may improve the cache hit rate as well as performance. For embedded systems, it can reduce the memory requirement, the memory latency and the energy consumption. This paper investigates program transformations which a compiler can use to reduce the memory space required for storing program data. In particular, the paper uses integer programming to model the problem of combining loop shifting, loop fusion and array contraction to minimize the data memory required to execute a collection of multi-level loop nests. The integer programming problem is then reduced to an equivalent network flow problem which can be solved in polynomial time.

show abstract

Section: Resultsmentioning

confidence: 99%

A Polynomial-Time Algorithm for Memory Space Reduction

Song

Wang

2005

Int J Parallel Prog

Self Cite

View full text Add to dashboard Cite

show abstract

“…An opposite approach is read-centric, corresponding to backward propagation. In a later experiment to actually build a compiler prototype, my student, Junjie Gu, and I took the latter approach [7].…”

Section: The Top-level Design Reexaminedmentioning

confidence: 99%

Author's retrospective for array privatization for parallel execution of loops

2014

25th Anniversary International Conference on Supercomputing Anniversary Volume -

Self Cite

View full text Add to dashboard Cite

Array privatization, as a program transform to increase the opportunity for loop parallelization, was introduced at a time when numerical programs dominated the parallel computing world. Today, with parallel processors becoming ubiquitous, the computer industry faces a new challenge, i.e. how to best utilize hardware parallelism for the next generation of main-stream applications. The author re-examines the technique of array privatization both in its historical context and in view of new developments in recent years.

show abstract

“…Gu et al [34] improved array region analysis by analyzing conditional array regions. They also describe array regions by a set of convex regions.…”

Section: Array Region Analysismentioning

confidence: 99%

A unified framework for nonlinear dependence testing and symbolic analysis

Engelen

Birch

Shou

et al. 2004

Proceedings of the 18th Annual International Conference on Supercomputing

View full text Add to dashboard Cite

This paper presents a unified approach for generalized induction variable recognition and substitution, pointer analysis, analysis of conditionally updated variables, value range analysis, array region analysis, and nonlinear dependence testing. The analysis techniques share a well-defined uniform approach based on the chains of recurrences algebra. The uniform algebraic approach provides a powerful unified framework for developing analysis algorithms for restructuring compilers. The paper introduces a new set of analysis algorithms that accurately handle conditional control flow, pointer arithmetic, and nonlinear symbolic expressions in loops, which are known to be problematic for conventional restructuring compilers.

show abstract

Experience with efficient array data flow analysis for array privatization

Cited by 30 publications

References 25 publications

A Polynomial-Time Algorithm for Memory Space Reduction

A Polynomial-Time Algorithm for Memory Space Reduction

Author's retrospective for array privatization for parallel execution of loops

A unified framework for nonlinear dependence testing and symbolic analysis

Contact Info

Product

Resources

About