Abstract | ||
---|---|---|
This paper examines a previously unanalyzed bistability phenomenon with respect to the number of threads that are doing useful work. This phenomenon is illustrated by a single work queue on a shared-memory machine. An analysis of designs that use two separate memory accesses to lock and unlock critical sections (split transaction) and that employ a first come/first serve queuing mechanism for shared-memory locations is presented. A bistability in the number of threads working, brought about by these conditions, is analyzed and experimentally demonstrated. A simple analysis is presented which predicts the throughput at a critical section of code as a function of the number of applied threads. The study concludes that the mean size of the work items that can be executed in parallel without the possibility of stalling is proportional to the square of the number of threads applied. |
Year | DOI | Venue |
---|---|---|
1993 | 10.1007/BF01206243 | The Journal of Supercomputing |
Keywords | Field | DocType |
Critical section,interconnection network,instruction thread,interprocess synchronization,MIMD,Markov chain,shared memory,single access lock | Bistability,Shared memory,Lock (computer science),Computer science,Critical section,Parallel computing,Queue,Thread (computing),Queueing theory,Distributed computing,MIMD | Journal |
Volume | Issue | ISSN |
7 | 3 | 0920-8542 |
Citations | PageRank | References |
2 | 0.42 | 7 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Raymond R. Glenn | 1 | 11 | 2.79 |
Daniel V. Pryor | 2 | 51 | 16.93 |
John M. Conroy | 3 | 2 | 0.42 |
Theodore Johnson | 4 | 2 | 0.42 |