Models of Multiprocessor Architectures and Algorithms
Abstract
A key requirement for the effective use of multiprocessors in real-world applications is an ability to accurately predict the performance of a specific algorithm on a specific architecture. This research developed such a prediction methodology, which permits separate evaluation of algorithm and architecture performance, with only a small number of 'cross' parameters required to link the models. Additional results include program behavior models that lead to effective trace compression techniques.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 25, 1997
- Accession Number
- ADA386857
Entities
People
- Eric E. Johnson
Organizations
- New Mexico State University