Double-Scan: Introducing and Implementing a New Data-Parallel Skeleton

Bischof, Holger; Gorlatch, Sergei

doi:10.1007/3-540-45706-2_88

Cited by 7 publications

(9 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [1], we introduced a more special skeleton, double-scan (DS), which is more convenient than DH for expressing some application classes: Definition 2 For binary operators © and <8>, two double-scan (DS) skeletons are defined:…”

Section: Skeletons and Their Cost Propertiesmentioning

confidence: 99%

“…Both double-scan skeletons have two functional parameters, which are the base operators of their constituent scans. If the parameter operator © is associative, then as shown in [1], DS can be parallelized. An important efficiency criterion for parallel algorithms is their cost, which is defined as the product of the required time and the number of processors used, i.e.…”

Section: Skeletons and Their Cost Propertiesmentioning

confidence: 99%

“…Our first question is whether the known implementations of DH [9] and DS [1] are cost-optimal. We are especially interested in a skeleton's generic implementation, which can be applied to all particular instances of the skeleton.…”

Section: Skeletons and Their Cost Propertiesmentioning

confidence: 99%

“…On the one hand, as proved in [1], the DS skeleton can be expressed as an instance of the DH skeleton for particular values of parameters © and ®, with additional preand postcomputations performed locally. On the other hand, the sequential time complexity of DS is obviously linear.…”

Section: Skeletons and Their Cost Propertiesmentioning

confidence: 99%

“…its time-processor product [17], is optimal, and 2) whether the performance of a skeleton-based program can be predicted in advance. We answer these questions for two fairly complicated skeletons introduced by ourselves earlier, both embodying the divide-and-conquer paradigm: DH (distributable homomorphism) [9] and DS (double-scan) [1].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Cost Optimality and Predictability of Parallel Programming With Skeletons

Bischof

Gorlatch

Kitzelmann

2003

Parallel Process. Lett.

Self Cite

View full text Add to dashboard Cite

Skeletons are reusable, parameterized program components with well-defined semantics and pre-packaged efficient parallel implementation. This paper develops a new, provably cost-optimal implementation of the DS (double-scan) skeleton for programming divide-and-conquer algorithms. Our implementation is based on a novel data structure called plist (pointed list); implementation's performance is estimated using an analytical model. We demonstrate the use of the DS skeleton for parallelizing a tridiagonal system solver and report experimental results for its MPI implementation on a Cray T3E and a Linux cluster: they confirm the performance improvement achieved by the cost-optimal implementation and demonstrate its good predictability by our performance model.

show abstract

Section: Skeletons and Their Cost Propertiesmentioning

confidence: 99%

Section: Skeletons and Their Cost Propertiesmentioning

confidence: 99%

Section: Skeletons and Their Cost Propertiesmentioning

confidence: 99%

Section: Skeletons and Their Cost Propertiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Cost Optimality and Predictability of Parallel Programming With Skeletons

Bischof

Gorlatch

Kitzelmann

2003

Parallel Process. Lett.

Self Cite

View full text Add to dashboard Cite

show abstract

High‐level specifications for automatically generating parallel code

Acosta

Almeida

Peláez

2012

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYThe arrival of multicore systems, along with the speed‐up potential available in graphics processing units, has given us unprecedented low‐cost computing power. These systems address some of the known architecture problems but at the expense of considerably increased programming complexity. Heterogeneity, at both the architectural and programming levels, poses a great challenge to programmers. Many proposals have been put forth to facilitate the job of programmers. Leaving aside proposals based on the development of new programming languages because of the effort this represents for the user (effort to learn and reuse code), the remaining proposals are based on transforming sequential code into parallel code, or on transforming parallel code designed for one architecture into parallel code designed for another. A different approach relies on the use of skeletons. The programmer has available set of parallel standards that comprise the basis for developing parallel code while programming sequential code. In this context, we propose a methodology for developing an automatic source‐to‐source transformation in a specific domain. This methodology is instantiated in a framework aimed at solving dynamic programming problems. Using this framework, the final user (a physician, mathematician, biologist, etc.) can express her problem using an equation in Latex, and the system will automatically generate the optimal parallel code for homogeneous or heterogeneous architectures. This approach allows for great portability toward these new emerging architectures and for great productivity, as evidenced by the computational results.Copyright © 2012 John Wiley & Sons, Ltd.

show abstract

Optimizing Sequences of Skeleton Calls

Kuchen

2004

Domain-Specific Program Generation

View full text Add to dashboard Cite

Double-Scan: Introducing and Implementing a New Data-Parallel Skeleton

Cited by 7 publications

References 5 publications

Cost Optimality and Predictability of Parallel Programming With Skeletons

Cost Optimality and Predictability of Parallel Programming With Skeletons

High‐level specifications for automatically generating parallel code

Optimizing Sequences of Skeleton Calls

Contact Info

Product

Resources

About