Multithreaded Value Prediction

Authors:
Nathan Tuck;Dean M. Tullsen
Affiliations:
University of California, San Diego;University of California, San Diego
Venue:
HPCA '05 Proceedings of the 11th International Symposium on High-Performance Computer Architecture
Year:
2005

Citing 0
Cited 8

Tolerating Dependences Between Large Speculative Threads Via Sub-Threads

Proceedings of the 33rd annual international symposium on Computer Architecture
CAVA: Using checkpoint-assisted value prediction to hide L2 misses

ACM Transactions on Architecture and Code Optimization (TACO)
Enabling real-time physics simulation in future interactive entertainment

Proceedings of the 2006 ACM SIGGRAPH symposium on Videogames
Memory-level parallelism aware fetch policies for simultaneous multithreading processors

ACM Transactions on Architecture and Code Optimization (TACO)
The potential of using dynamic information flow analysis in data value prediction

Proceedings of the 19th international conference on Parallel architectures and compilation techniques
Energy-performance design space exploration in SMT architectures exploiting selective load value predictions

Proceedings of the Conference on Design, Automation and Test in Europe
Leveraging Strength-Based Dynamic Information Flow Analysis to Enhance Data Value Prediction

ACM Transactions on Architecture and Code Optimization (TACO)
A bypass mechanism to enhance branch predictor for SMT processors

ACSAC'07 Proceedings of the 12th Asia-Pacific conference on Advances in Computer Systems Architecture

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper introduces a novel technique which leverages value prediction and multithreading on a simultaneous multithreading processor to achieve higher performance in a single threaded application. By allowing the value-speculative execution to proceed in a separate thread, this technique overcomes barriers that make traditional value prediction relatively ineffective for tolerating long latency loads. It shows that this technique can be as much as 2-5 times more effective than traditional value prediction, achieving more than 40% average performance gain on the SPEC benchmarks with realistic hardware parameters. These gains come from two effects: allowing greater separation between the stalled load and the speculative execution, and the ability to speculate on multiple values for a single load.