Distributed scheduling algorithms to improve the performance of parallel data transfers

Authors:
Dannie Durand;Ravi Jain;David Tseytlin
Affiliations:
Bellcore, 445 South Street, Morristown, NJ;Bellcore, 445 South Street, Morristown, NJ;Bellcore, 445 South Street, Morristown, NJ
Venue:
ACM SIGARCH Computer Architecture News - Special issue on input/output in parallel computer systems
Year:
1994

Citing 3
Cited 1

Fast randomized algorithms for distributed edge coloring

PODC '92 Proceedings of the eleventh annual ACM symposium on Principles of distributed computing
Efficient optical communication in parallel computers

SPAA '92 Proceedings of the fourth annual ACM symposium on Parallel algorithms and architectures
High-speed switch scheduling for local-area networks

ACM Transactions on Computer Systems (TOCS)

An experimental study of a simple, distributed edge-coloring algorithm

Journal of Experimental Algorithmics (JEA)

Quantified Score

Hi-index	0.00

Visualization

Abstract

The cost of data transfers, and in particular of I/O operations, is a growing problem in parallel computing. A promising approach to alleviating this bottleneck is to schedule parallel I/O operations explicitly. We develop a class of decentralized algorithms for scheduling parallel I/O operations, where the objective is to reduce the time required to complete a given set of transfers. These algorithms, based on edge-coloring and matching of bipartite graphs, rely upon simple heuristics to obtain shorter schedules. We present simulation results indicating that the best of our algorithms can produce schedules whose length is within 2--20% of the optimal schedule, a substantial improvement on previous decentralized algorithms. We discuss theoretical and experimental work in progress and possible extensions.