Putting Data Value Predictors to Work in Fine-Grain Parallel Processors
HiPC '01 Proceedings of the 8th International Conference on High Performance Computing
Using Dataflow Based Contextfor Accurate Branch Prediction
HiPC '02 Proceedings of the 9th International Conference on High Performance Computing
Detecting global stride locality in value streams
Proceedings of the 30th annual international symposium on Computer architecture
The significance of affectors and affectees correlations for branch prediction
HiPEAC'08 Proceedings of the 3rd international conference on High performance embedded architectures and compilers
Improving branch prediction by considering affectors and affectees correlations
Transactions on high-performance embedded architectures and compilers III
Neural confidence estimation for more accurate value prediction
HiPC'05 Proceedings of the 12th international conference on High Performance Computing
Leveraging Strength-Based Dynamic Information Flow Analysis to Enhance Data Value Prediction
ACM Transactions on Architecture and Code Optimization (TACO)
Exploiting thread-level speculative parallelism with software value prediction
ACSAC'05 Proceedings of the 10th Asia-Pacific conference on Advances in Computer Systems Architecture
| Hi-index | 0.00 |
Abstract: We explore the reasons behind the rather low prediction accuracy of existing data value predictors. Our studies show that contexts formed only from the outcomes of the last several instances of a static instruction do not always encapsulate all of the information required for correct prediction. Complex interactions between data flow and control flow change the context in ways that result in predictability loss for a significant number of dynamic instructions. For improving the prediction accuracy, we propose the concept of using contexts derived from the predictable portions of the data flow graph. That is, the predictability of hard-to-predict instructions can be improved by taking advantage of the predictability of the easy-to-predict instructions that precede it in the data flow graph. We propose and investigate a run-time scheme for producing such an improved context from the predicted values of previous instructions. We also propose a novel predictor called dynamic data flow-inherited speculative context (DDISC)based predictor for specifically predicting hard-to-predict instructions. Simulation results verify that the use of data flow-based contexts yields significant improvements in prediction accuracies, ranging from 35%to 99%.This translates to an overall prediction accuracy of 68%to 99.9%.