Collecting unused processing capacity: an analysis of transient distributed systems
- 7 January 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 482-489
- https://doi.org/10.1109/icdcs.1989.37980
Abstract
Distributed systems having large numbers of idle computers and workstations are analyzed using a very simple model of a distributed program (a fixed amount of work) to see how the use of transient processors affects the program's service time. The probability density of the length of time it takes to finish a fixed amount of work is determined. An equation is given for the main result for an M-processor network. Simulations confirm that Brownian motion with drift is an accurate model of system performance. With large programs that run for a long time relative to the length of available and nonavailable periods, the central limit-theorem applies, and the Brownian-motion-with-drift model remains good regardless of the distributions of the available and the nonavailable periods. Under these assumptions, the distribution of finishing time is very tight about its mean and well approximated by a normal distribution.Keywords
This publication has 5 references indexed in Scilit:
- Calculating availability and performability measures of repairable computer systems using randomizationJournal of the ACM, 1989
- The Benevolent Bandit Laboratory: a testbed for distributed algorithmsIEEE Journal on Selected Areas in Communications, 1989
- Closed-Form Solution for System Availability DistributionIEEE Transactions on Reliability, 1987
- Using idle workstations in a shared computing environmentPublished by Association for Computing Machinery (ACM) ,1987
- Distributed systemsCommunications of the ACM, 1985