The data alignment phase in compiling programs for distributed-memory machines

Li, Jingke; Chen, Marina

doi:10.1016/0743-7315(91)90090-v

Cited by 113 publications

(66 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A do edge with dashed arrow line is represented by the loop structure linked from outer loop i to inner loop j. We know that the data alignment problem is an NP-complete problem [2,12]. However, in this paper, we use a new kind of graphs to capture the characteristics of a program and data arrays as well as intent to optimize the data communication overhead.…”

Section: Skewed Array Alignmentmentioning

confidence: 99%

“…Note that Li and Chen's CAG [12] for Cholesky factorization algorithm has no edge because these is only a single array A in this loop nest. Hence, it is impossible to determine the kind of alignment to a certain data array which is good.…”

Section: Alignment For a Data Arraymentioning

confidence: 99%

“…After that, we have to take into account how to determine data alignment with these multiple data arrays accessed by the program. Here we first transform the CCAG derived from the previous subsection into a CAG (component affinity graph) in [12]. Next, the proposed alignment algorithm [12] used here is applied to the CAG to solve our alignment problem.…”

Section: Aligning Multiple Data Arrays For a Programmentioning

confidence: 99%

“…Li and Chen [12] studied the problem of aligning data arrays in order to minimize communications. Meanwhile, they also showed the alignment problem is NP-complete in the number of data arrays with a loop nest.…”

Section: Introductionmentioning

confidence: 99%

“…Meanwhile, they also showed the alignment problem is NP-complete in the number of data arrays with a loop nest. Gupta and Banerjee [8] used the constraint-based method to extend the alignment method [12] for the distributed memory multicomputers. Lee [10] solved the data redistribution problem among loop nests using a dynamic programming technique based on Gupta and Banerjee's method [8].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Skewed Data Partition and Alignment Techniques for Compiling Programs on Distributed Memory Multicomputers

Chen

Chang

2000

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Minimizing data communication over processors is the key to compile programs for distributed memory multicomputers. In this paper, we propose new data partition and alignment techniques for partitioning and aligning data arrays with a program in a way of minimizing communication over processors. We use skewed alignment instead of the dimension-ordered alignment techniques to align data arrays. By developing the skewed scheme, we can solve more complex programs with minimized data communication than that of the dimension-ordered scheme. Finally, we compare the proposed scheme with the dimension-ordered alignment one by experimental results. The experimental results show that our proposed scheme has more opportunities to align data arrays such that data communications over processors can be minimized.

show abstract

Section: Skewed Array Alignmentmentioning

confidence: 99%

Section: Alignment For a Data Arraymentioning

confidence: 99%

Section: Aligning Multiple Data Arrays For a Programmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Skewed Data Partition and Alignment Techniques for Compiling Programs on Distributed Memory Multicomputers

Chen

Chang

2000

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

CRAFT: a framework for F90/HPF compiler optimizations

Chen

Cowie

1999

Concurrency: Pract. Exper.

Self Cite

View full text Add to dashboard Cite

In this paper, we give an overview of the results of the CRAFT optimising compiler project (Fortran 90/HPF subset compilers). We start by describing the theoretical framework within which we designed program transformations for the optimization of inter‐ and intra‐procedural data motion, as well as the optimizations for parallel loops; we then describe the implementation of the CRAFT compilers for Thinking Machines' CM‐2 and CM‐5. We report results from experiments on the Connection Machine CM‐5, the IBM SP‐2 and a network of UltraSparc workstations. The results demonstrate that these optimizations can achieve significant object code performance improvement. Copyright © 1999 John Wiley & Sons, Ltd.

show abstract

Experiments with Program unification on the Cray Y‐MP

Chuang¹,

Rego²,

Mathur³

1994

Concurrency: Pract. Exper.

View full text Add to dashboard Cite

Program unification is a technique for source‐to‐source transformation of code for enhanced execution performance on vector and SIMD architectures. This work focuses on simple examples of program unification to explain the methodology and demonstrate its promise as a practical technique for improved performance. Using simple examples to explain how unification is done, we outline two experiments in the simulation domain that benefit from unification, namely Monte Carlo and discrete‐event simulation. Empirical tests of unified code on a Cray Y‐MP multiprocessor show that unification improves execution performance by a factor of roughly 8 for given application. The technique is general in that it can be applied to computation‐intensive programs in various data‐parallel application domains.

show abstract

The data alignment phase in compiling programs for distributed-memory machines

Cited by 113 publications

References 5 publications

Skewed Data Partition and Alignment Techniques for Compiling Programs on Distributed Memory Multicomputers

Skewed Data Partition and Alignment Techniques for Compiling Programs on Distributed Memory Multicomputers

CRAFT: a framework for F90/HPF compiler optimizations

Experiments with Program unification on the Cray Y‐MP

Contact Info

Product

Resources

About