Hybrid analytical-statistical modeling for efficiently exploring architecture and workload design spaces

Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT

Volumn , Issue , 2001, Pages 25-34

  Eeckhout, L ^a   De Bosschere, K ^a  

^a GHENT UNIVERSITY   (Belgium)

Author keywords

[No Author keywords available]

Indexed keywords

COMPUTER ARCHITECTURE; COMPUTER SIMULATION; PROGRAM COMPILERS; STATISTICAL METHODS;

MICROPROCESSOR DESIGN;

MICROPROCESSOR CHIPS;

EID: 0035176094     PISSN: 1089795X     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (15)

1. Performance analysis and its impact on design
2. Modeling superscalar processors via statistical simulation
3. Instruction window size trade-offs and characterization of program parallelism
4. Increasing the accuracy of statistical simulation for modeling superscalar processors
5. Performance analysis through synthetic trace generation
6. Register traffic analysis for streamlining inter-operation communication in fine-grain parallel processors
7. Micro-processor power estimation using profile-driven program synthesis
8. Workload characterization: Motivation, goals and methodology
9. Dynamic trace analysis for analytic modeling of superscalar performance
10. Theoretical modeling of superscalar processor performance
11. A framework for statistical modeling of superscalar processor performance
12. HLS: Combining statistical and symbolic simulation to guide microprocessor design
13. Analytic evaluation of shared-memory systems with ILP processors
14. Instruction fetching: Coping with code bloat