Tradeoffs in Buffering Speculative Memory State for Thread-Level Speculation in Multiprocessors (2005)
Cached
Download Links
- [polaris.cs.uiuc.edu]
- [iacoma.cs.uiuc.edu]
- [polaris.cs.uiuc.edu]
- [webdiis.unizar.es]
- DBLP
Other Repositories/Bibliography
| Venue: | ACM Transactions on Architecture and Code Optimization |
| Citations: | 21 - 3 self |
BibTeX
@ARTICLE{Garzaran05tradeoffsin,
author = {Maria Jesus Garzaran and Milos Prvulovic and Jose Maria Llaberia},
title = {Tradeoffs in Buffering Speculative Memory State for Thread-Level Speculation in Multiprocessors},
journal = {ACM Transactions on Architecture and Code Optimization},
year = {2005},
volume = {2},
pages = {2005}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
this paper, we introduce a novel taxonomy of approaches to buffer and manage multiversion speculative memory state in multiprocessors. We also present a detailed complexity-benefit tradeoff analysis of the different approaches. Finally, we use numerical applications to evaluate the performance of the approaches under a single architectural framework. Our key insights are that support for buffering the state of multiple speculative tasks and versions per processor is more complexity-effective than support This paper extends an earlier version that appeared in the 9th International Symposium on High Performance Computer Architecture (HPCA), February 2003
Keyphrases
thread-level speculation buffering speculative memory state different approach high performance computer architecture single architectural framework novel taxonomy multiversion speculative memory state international symposium detailed complexity-benefit tradeoff analysis key insight multiple speculative task numerical application
