Name
Title | ||
|---|---|---|
Architectural support and mechanisms for object caching in dynamic multithreaded computations |
Abstract | ||
|---|---|---|
High-level parallel programming models supporting dynamic fine-grained threads in a global object space are becoming increasingly popular for expressing irregular applications based on sophisticated adaptive algorithms and pointer-based data structures. However, implementing these multithreaded com- putations on scalable parallel machines poses significant challenges, particularly with respect to object caching. Object caching techniques must be able to tolerate unresponsive processors and protocol han- dler occupancy delays. This paper examines whether these challenges can be offset by leveraging responsive general-purpose communication architectural features (such as remote memory access and atomic operations), possibly compensating for the lack of more sophisticated hardware primitives by relying upon increased involve- ment of the run-time system and the compiler. A detailed performance analysis of four irregular appli- cations, using the Illinois Concert System on the Cray T3D and the SGI Origin 2000, finds that existing software distributed shared memory (DSM) systems are capable of delivering good performance only in the presence of a high level of responsive communication architecture support (specifically, support for remote atomic operations). Recognizing that this situation stems from the synchronous request-reply nature of DSM protocols, we present a composable object caching framework, called view caching, which exploits knowledge of application data access semantics to construct custom protocols that re- quire reduced processor synchronization. View caching protocols are more tolerant to responsiveness and occupancy delays and are able to exploit even lower-level responsive communication primitives (such as non-atomic remote memory accesses) for a performance benefit. |
Year | DOI | Venue |
|---|---|---|
1999 | 10.1006/jpdc.1999.1555 | J. Parallel Distrib. Comput. |
Keywords | Field | DocType |
architectural support,dynamic multithreaded computation,data structure,data access,parallel programming model | Data structure,Synchronization,Parallel algorithm,CPU cache,Computer science,Parallel computing,Thread (computing),Compiler,Data access,Distributed computing,Scalability | Journal |
Volume | Issue | ISSN |
58 | 2 | Journal of Parallel and Distributed Computing |
Citations | PageRank | References |
1 | 0.36 | 31 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Vijay Karamcheti | 1 | 646 | 67.03 |
| Andrew A. Chien | 2 | 3696 | 405.97 |