Skeletons for divide and conquer algorithms

Authors:
Michael Poldner;Herbert Kuchen
Affiliations:
University of Münster, Münster, Germany;University of Münster, Münster, Germany
Venue:
PDCN '08 Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Networks
Year:
2008

Citing 13
Cited 1

Algorithmic skeletons: structured management of parallel computation

Algorithmic skeletons: structured management of parallel computation
Parallel skeletons for structured composition

PPOPP '95 Proceedings of the fifth ACM SIGPLAN symposium on Principles and practice of parallel programming
N-graphs: scalable topology and design of balanced divide-and-conquer algorithms

Parallel Computing
Efficient load balancing for wide-area divide-and-conquer applications

PPoPP '01 Proceedings of the eighth ACM SIGPLAN symposium on Principles and practices of parallel programming
On Dividing and Conquering Independently

Euro-Par '97 Proceedings of the Third International Euro-Par Conference on Parallel Processing
A Skeleton Library

Euro-Par '02 Proceedings of the 8th International Euro-Par Conference on Parallel Processing
MALLBA: A Library of Skeletons for Combinatorial Optimisation (Research Note)

Euro-Par '02 Proceedings of the 8th International Euro-Par Conference on Parallel Processing
Efficient Parallel Programming with Algorithmic Skeletons

Euro-Par '96 Proceedings of the Second International Euro-Par Conference on Parallel Processing - Volume I
Task and data parallelism in P3L

Patterns and skeletons for parallel and distributed computing
Bringing skeletons out of the closet: a pragmatic manifesto for skeletal parallel programming

Parallel Computing
A library of constructive skeletons for sequential style of parallel programming

InfoScale '06 Proceedings of the 1st international conference on Scalable information systems
A parallel symmetric block-tridiagonal divide-and-conquer algorithm

ACM Transactions on Mathematical Software (TOMS)
Flexible skeletal programming with eskel

Euro-Par'05 Proceedings of the 11th international Euro-Par conference on Parallel Processing

Algorithmic skeletons for multi-core, multi-GPU systems and clusters

International Journal of High Performance Computing and Networking

Quantified Score

Hi-index	0.00

Visualization

Abstract

Algorithmic skeletons intend to simplify parallel programming by providing recurring forms of program structure as predefined components. We present a fully distributed task parallel skeleton for a very general class of divide and conquer algorithms for MIMD machines with distributed memory. This approach is compared to a simple master-worker design. Based on experimental results for different example applications such as Mergesort, the Karatsuba multiplication algorithm and Strassen's algorithm for matrix multiplication, we show that the distributed workpool enables good runtimes and in particular scalability. Moreover, we discuss some implementation aspects for the distributed skeleton, such as the underlying data structures and load balancing strategy, in detail.