Piecewise Execution of Nested Parallel Programs - A Thread-Based Approach

Pfannenstiel, W.

doi:10.1007/3-540-48311-x_59

Cited by 2 publications

(1 citation statement)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this case, arrays that are not required for the current calculation could be evacuated from the device by transferring data back to the associated host-side array, and re-transferred back to the device only once required. Further inspiration for reducing the intermediate memory footprint of computations could be found in the areas of stream processing [131] or the piecewise execution of nested data-parallel programs [66,79,106].…”

Section: Discussionmentioning

confidence: 99%

Optimising purely functional GPU programs

McDonell

Chakravarty

Keller

et al. 2013

Proceedings of the 18th ACM SIGPLAN International Conference on Functional Programming

View full text Add to dashboard Cite

Purely functional, embedded array programs are a good match for SIMD hardware, such as GPUs. However, the naive compilation of such programs quickly leads to both code explosion and an excessive use of intermediate data structures. The resulting slowdown is not acceptable on target hardware that is usually chosen to achieve high performance. In this paper, we discuss two optimisation techniques, sharing recovery and array fusion, that tackle code explosion and eliminate superfluous intermediate structures. Both techniques are well known from other contexts, but they present unique challenges for an embedded language compiled for execution on a GPU. We present novel methods for implementing sharing recovery and array fusion, and demonstrate their effectiveness on a set of benchmarks.

show abstract

Section: Discussionmentioning

confidence: 99%

Optimising purely functional GPU programs

McDonell

Chakravarty

Keller

et al. 2013

Proceedings of the 18th ACM SIGPLAN International Conference on Functional Programming

View full text Add to dashboard Cite

show abstract

Combining Fusion Optimizations and Piecewise Execution of Nested Data-Parallel Programs

Pfannenstiel

2000

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Nested data-parallel programs often have large memory requirements due to their high degree of parallelism. Piecewise execution is an implementation technique used to minimize the space needed. In this paper, we present a combinination of piecewise execution and loop-fusion techniques. Both a formal framework and the execution model based on threads are presented. We give some experimental results, which demonstrate the good performance in memory consumption and execution time.

show abstract

Piecewise Execution of Nested Parallel Programs - A Thread-Based Approach

Cited by 2 publications

References 4 publications

Optimising purely functional GPU programs

Optimising purely functional GPU programs

Combining Fusion Optimizations and Piecewise Execution of Nested Data-Parallel Programs

Contact Info

Product

Resources

About