Mobile pipelines: parallelizing left-looking algorithms using navigational programming

Authors:
Lei Pan;Ming Kin Lai;Michael B. Dillencourt;Lubomir F. Bic
Affiliations:
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA;Donald Bren School of Information & Computer Sciences, University of California, Irvine, CA;Donald Bren School of Information & Computer Sciences, University of California, Irvine, CA;Donald Bren School of Information & Computer Sciences, University of California, Irvine, CA
Venue:
HiPC'05 Proceedings of the 12th international conference on High Performance Computing
Year:
2005

Citing 8
Cited 1

ScaLAPACK user's guide

ScaLAPACK user's guide
Loop tiling for parallelism

Loop tiling for parallelism
Solving Linear Systems on Vector and Shared Memory Computers

Solving Linear Systems on Vector and Shared Memory Computers
Automatic State Capture of Self-Migrating Computations in MESSENGERS

MA '98 Proceedings of the Second International Workshop on Mobile Agents
Distributed Sequential Computing Using Mobile Code: Moving Computation to Data

ICPP '01 Proceedings of the International Conference on Parallel Processing
Parallel Programming: Techniques and Applications Using Networked Workstations and Parallel Computers (2nd Edition)

Parallel Programming: Techniques and Applications Using Networked Workstations and Parallel Computers (2nd Edition)
Distributed parallel computing using navigational programming

International Journal of Parallel Programming
Look left, look right, look left again: an application of fractal symbolic analysis to linear algebra code restructuring

International Journal of Parallel Programming

Mobile agents, DSM, coordination, and self-migrating threads: a Common Framework

DNCOCO'08 Proceedings of the 7th conference on Data networks, communications, computers

Quantified Score

Hi-index	0.00

Visualization

Abstract

We consider the class of “left-looking” sequential matrix algorithms: consumer-driven algorithms that are characterized by “lazy” propagation of data. Left-looking algorithms are difficult to parallelize using the message-passing or distributed shared memory models because they only possess pipeline parallelism. We show that these algorithms can be directly parallelized using mobile pipelines provided by the Navigational Programming methodology. We present performance data demonstrating the effectiveness of our approach.