CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Efficient Partial Roll-backing Mechanism for Transactional Memory Systems

Mrida Mohammad Waliullah (Institutionen för data- och informationsteknik, Datorteknik (Chalmers))
Lecture Notes in Computer Science. Transactions on High-Performance Embedded Architectures and Compilers (0302-9743). Vol. 6590 (2008), p. 256-274.
[Artikel, refereegranskad vetenskaplig]

Transactional memory systems promise to reduce the burden of exposing thread-level parallelism in programs by relieving programmers from analyzing complex inter-thread dependences in detail. By encapsulating large program code blocks and executing them as atomic blocks, dependence checking is deferred to run-time. One of many conflicting transactions will then be committed whereas the others will have to roll-back and re-execute. In current proposals, a checkpoint is taken at the beginning of the atomic block and all execution can be wasted even if the conflicting access happens at the end of the atomic block. In this paper, we propose a novel scheme that (1) predicts when the first conflicting access occurs and (2) inserts a checkpoint before it is executed. When the prediction is correct, the only execution discarded is the one that has to be re-done. When the prediction is incorrect, the whole transaction has to be re-executed just as before. Overall, we find that our scheme manages to maintain high prediction accuracy and leads to a quite significant reduction in the number of lost cycles due to roll-backs; the geometric mean speedup across five applications is 16%.

Nyckelord: Transactional Memory, Intermediate checkpointing



Denna post skapades 2009-12-17. Senast ändrad 2012-02-14.
CPL Pubid: 104019

 

Läs direkt!


Länk till annan sajt (kan kräva inloggning)


Institutioner (Chalmers)

Institutionen för data- och informationsteknik, Datorteknik (Chalmers)

Ämnesområden

Datorteknik

Chalmers infrastruktur

Relaterade publikationer

Denna publikation ingår i:


Techniques to Reduce Inefficiencies in Hardware Transactional Memory Systems