Applying randomized edge coloring algorithms to distributed communication: an experimental study

We propose a parameterized, randomized edge coloring algorithm for use in coordinating data transfers in fully connected distributed architectures such as parallel 1/0 subsystems and multimedia information systems. Our approach is to preschedule 1/0 requests to eliminate contention for 1/0 ports while maintaining an efficient use of bandwidth. Request scheduling is equivalent to edge coloring a bipartite graph representing pending 1/0 requests. Although efficient optimal algorithms exist for centralized edge coloring where the global request graph is known, in distributed architectures heuristics must be used. We propose such heuristics and use experimental analysis to determine their ability to approach the centralized optimal. The performance of our algorithms is also compared with the work of other researchers experimentally. Our results show that our algorithms produce schedules within 5% of the optimal schedule, a substantial improvement over existing algorithms. The use of experimental analysis allows us to evaluate the appropriateness of each heuristic for a variety of different architectural models and applications.

[1]  Gordon Bell,et al.  C.mmp: a multi-mini-processor , 1972, AFIPS '72 (Fall, part II).

[2]  Dannie Durand,et al.  Impact of Memory Contention on Dynamic Scheduling on Numa Multiprocessors , 1993, 1993 International Conference on Parallel Processing - ICPP'93.

[3]  Ravi Jain,et al.  Scheduling Parallel I/O Operations in Multiple Bus Systems , 1992, J. Parallel Distributed Comput..

[4]  Leslie Ann Goldberg,et al.  An Ω(√ log log n) lower bound for routing in optical networks , 1994, SPAA '94.

[5]  Proceedings of the Fall Joint Computer Conference, November 2-6, 1986, Dallas, Texas, USA , 1986, Fall Joint Computer Conference.

[6]  Thomas H. Cormen Fast Permuting on Disk Arrays , 1993, J. Parallel Distributed Comput..

[7]  Richard J. Anderson Optical Communication for Pointer Based Algorithms , 1988 .

[8]  Jan Karel Lenstra,et al.  Recent developments in deterministic sequencing and scheduling: a survey : (preprint) , 1981 .

[9]  Ravi Jain Scheduling data transfers in parallel computers and communications systems , 1992 .

[10]  Michael Luby Removing randomness in parallel computation without a processor penalty , 1988, [Proceedings 1988] 29th Annual Symposium on Foundations of Computer Science.

[11]  Thanasis Tsantilas,et al.  Efficient optical communication in parallel computers , 1992, SPAA '92.

[12]  Thomas E. Anderson,et al.  High-speed switch scheduling for local-area networks , 1993, TOCS.

[13]  Edward G. Coffman,et al.  Computer and job-shop scheduling theory , 1976 .

[14]  Leslie G. Valiant,et al.  General Purpose Parallel Architectures , 1991, Handbook of Theoretical Computer Science, Volume A: Algorithms and Complexity.

[15]  Frank Thomson Leighton,et al.  A doubly logarithmic communication algorithm for the completely connected optical communication parallel computer , 1993, SPAA '93.

[16]  Mario J. Gonzalez Deterministic Processor Scheduling , 1977, CSUR.

[17]  Aravind Srinivasan,et al.  Fast randomized algorithms for distributed edge coloring , 1992, PODC '92.

[18]  Krishna Ssv. Palem On the Complexity of Precedence Constrained Scheduling , 1986 .

[19]  Aravind Srinivasan,et al.  Randomized Distributed Edge Coloring via an Extension of the Chernoff-Hoeffding Bounds , 1997, SIAM J. Comput..