DMCA
Feature selection and policy optimization for distributed instruction placement using reinforcement learning (2008)
Cached
Download Links
| Venue: | in The 17th International Conference on Parallel Architectures and Compilation Techniques |
| Citations: | 9 - 5 self |
BibTeX
@INPROCEEDINGS{Coons08featureselection,
author = {Katherine E. Coons and Behnam Robatmili and Matthew E. Taylor and A. Maher and Doug Burger and Kathryn S. Mckinley},
title = {Feature selection and policy optimization for distributed instruction placement using reinforcement learning},
booktitle = {in The 17th International Conference on Parallel Architectures and Compilation Techniques},
year = {2008}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Communication overheads are one of the fundamental challenges in a multiprocessor system. As the number of processors on a chip increases, communication overheads and the distribution of computation and data become increasingly important performance factors. Explicit Dataflow Graph Execution (EDGE) processors, in which instructions communicate with one another directly on a distributed substrate, give the compiler control over communication overheads at a fine granularity. Prior work shows that compilers can effectively reduce fine-grained communication overheads in EDGE architectures using a spatial instruction placement algorithm with a heuristic-based cost function. While this algorithm is effective, the cost function must be painstakingly tuned. Heuristics tuned to perform well across a variety of applications leave users with little ability to tune
Keyphrases
communication overhead reinforcement learning distributed instruction placement feature selection policy optimization fine-grained communication overhead spatial instruction placement algorithm heuristic-based cost function prior work show compiler control edge architecture little ability important performance factor distributed substrate cost function fundamental challenge multiprocessor system explicit dataflow graph execution fine granularity
