Title | ||
---|---|---|
Integrating non-blocking synchronisation in parallel applications: performance advantages and methodologies |
Abstract | ||
---|---|---|
In this paper we investigate how performance and speedup of applications would be affected by using non-blocking rather than blocking synchronisation in parallel systems. The results obtained show that for many applications, non-blocking synchronisation lead to significant speedups for a fairly large number of processors, while it never slows the applications down. As part of this investigation this paper also provides a set of efficient and simple translations that show how typical blocking operations found in parallel applications, such as simple locks, queues and lock trees can be translated into non-blocking equivalents that use hardware primitives common in modern multiprocessor systems. With these translations this paper clearly demonstrates that it is easy for the application designer/programmer to replace the blocking operations commonly found on with non-blocking equivalents ones. For the empirical results a set of representative applications running on a large-scale ccNUMA machine were used. |
Year | DOI | Venue |
---|---|---|
2002 | 10.1145/584369.584378 | Workshop on Software and Performance 2005 |
Keywords | Field | DocType |
empirical result,simple translation,parallel application,non-blocking equivalent,parallel system,large number,performance advantage,non-blocking synchronisation,simple lock,application designer,synchronisation lead,large-scale ccnuma machine,modeling,compositionality,parallel systems,uml | Principle of compositionality,Synchronization,Programmer,Unified Modeling Language,Lock (computer science),Computer science,Queue,Multiprocessing,Real-time computing,Speedup | Conference |
ISBN | Citations | PageRank |
1-58113-563-7 | 17 | 0.94 |
References | Authors | |
19 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Philippas Tsigas | 1 | 1200 | 99.58 |
Yi Zhang | 2 | 154 | 10.62 |