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

An All-Software Thread-Level Data Dependence Speculation System for Multiprocessors

Peter Rundberg (Institutionen för datorteknik) ; Per Stenström (Institutionen för datorteknik)
Journal of Instruction-Level Parallelism (1942-9525). Vol. 3 (2002),
[Artikel, refereegranskad vetenskaplig]

We present a software approach to design a thread-level data dependence speculation system targeting multiprocessors. Highly-tuned checking codes are associated with loads and stores whose addresses cannot be disambiguated by parallel compilers and that can potentially cause dependence violations at run-time. Besides resolving many name and true data dependencies through dynamic renaming and forwarding, respectively, our method supports parallel commit operations. Performance results collected on an architectural simulator and validated on a commercial multi-processor show that the overhead can be reduced to less than ten instructions per speculative memory operation. Moreover, we demonstrate that a ten-fold speedup is possible on some of the difficult-to-parallelize loops in the Perfect Club benchmark suite on a 16-way multiprocessor.

Nyckelord: Multiprocessors, thread-level speculation, parallel compilers, performance evaluation

Denna post skapades 2013-07-09. Senast ändrad 2013-07-09.
CPL Pubid: 179996


Läs direkt!

Länk till annan sajt (kan kräva inloggning)

Institutioner (Chalmers)

Institutionen för datorteknik (2002-2004)


Data- och informationsvetenskap

Chalmers infrastruktur