DMCA
Mitosis Compiler: An Infrastructure for Speculative Threading Based on PreComputation Slices (2005)
| Venue: | In Conference on Programming Language Design and Implementation |
| Citations: | 81 - 4 self |
BibTeX
@INPROCEEDINGS{Quiñones05mitosiscompiler:,
author = {Carlos García Quiñones and Carlos Madriles and Jesús Sánchez and Pedro Marcuello and Antonio González and Dean M. Tullsen},
title = {Mitosis Compiler: An Infrastructure for Speculative Threading Based on PreComputation Slices},
booktitle = {In Conference on Programming Language Design and Implementation},
year = {2005}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Speculative parallelization can provide significant sources of additional thread-level parallelism, especially for irregular applications that are hard to parallelize by conventional approaches. In this paper, we present the Mitosis compiler, which partitions applications into speculative threads, with special emphasis on applications for which conventional parallelizing approaches fail. The management of inter-thread data dependences is crucial for the performance of the system. The Mitosis framework uses a pure software approach to predict/compute the thread’s input values. This software approach is based on the use of pre-computation slices (p-slices), which are built by the Mitosis compiler and added at the beginning of the speculative thread. P-slices must compute thread input values accurately but they do not need to guarantee correctness, since the underlying architecture can detect and recover from misspeculations. This allows the compiler to use aggressive/unsafe optimizations to significantly reduce their overhead. The most important optimizations included in the Mitosis compiler and presented in this paper are branch pruning, memory and register dependence speculation, and early thread squashing. Performance evaluation of Mitosis compiler/architecture shows an average speedup of 2.2.
Keyphrases
mitosis compiler precomputation slice speculative threading speculative thread thread input value branch pruning register dependence speculation pre-computation slice significant source conventional approach aggressive unsafe optimization underlying architecture important optimization average speedup mitosis compiler architecture inter-thread data dependence conventional parallelizing approach software approach thread squashing performance evaluation additional thread-level parallelism speculative parallelization irregular application special emphasis mitosis framework pure software approach
