Title | ||
---|---|---|
DTM: Decoupled Hardware Transactional Memory to Support Unbounded Transaction and Operating System |
Abstract | ||
---|---|---|
Supporting unbounded transactions and operating system (OS) are two notable challenges that need to be efficiently resolved by practically accepted hardware transaction memory (HTM) systems. Current proposals that heavily rely on traditional cache system to handle version management or conflict detection support poorly to resolve these two challenges. Traditional approach brings either unavoidable design complexity or large performance costs. We propose DTM (decoupled transactional memory), a comprehensive hardware-based solution which follows a new design approach that fully decouples transaction processing from traditional cache system to resolve these two challenges. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ICPP.2009.52 | ICPP |
Keywords | Field | DocType |
unbounded transaction,decoupled transactional memory,cache system,cache storage,version management,design complexity,operating systems (computers),transaction processing,traditional cache system,operating system,comprehensive hardware-based solution,new design approach,unavoidable design complexity,decoupled hardware transactional memory,decouples transaction processing,conflict detection,accepted hardware transaction memory,traditional approach,decoupled hardware transactional memory systems,conflict detection support,coherence,hardware,kernel,protocols,hardware transactional memory,transactional memory | Transaction processing,Kernel (linear algebra),Permission,Yarn,Version management,Cache,Computer science,Parallel computing,Transactional memory,Database transaction,Operating system,Distributed computing | Conference |
ISSN | ISBN | Citations |
0190-3918 E-ISBN : 978-0-7695-3802-0 | 978-0-7695-3802-0 | 1 |
PageRank | References | Authors |
0.35 | 9 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shaogang Wang | 1 | 14 | 5.95 |
Weixia Xu | 2 | 64 | 16.66 |
Zhengbin Pang | 3 | 49 | 11.18 |
Dan Wu | 4 | 2318 | 272.22 |
Qiang Dou | 5 | 55 | 19.24 |
Xiaodong Yang | 6 | 5 | 6.51 |