Simultaneous multithreading: maximizing on-chip parallelism
ISCA '95 Proceedings of the 22nd annual international symposium on Computer architecture
ISCA '96 Proceedings of the 23rd annual international symposium on Computer architecture
The MIPS R10000 Superscalar Microprocessor
IEEE Micro
Front-End Policies for Improved Issue Efficiency in SMT Processors
HPCA '03 Proceedings of the 9th International Symposium on High-Performance Computer Architecture
The Impact of Resource Partitioning on SMT Processors
Proceedings of the 12th International Conference on Parallel Architectures and Compilation Techniques
Back-end assignment schemes for clustered multithreaded processors
Proceedings of the 18th annual international conference on Supercomputing
Dynamically Controlled Resource Allocation in SMT Processors
Proceedings of the 37th annual IEEE/ACM International Symposium on Microarchitecture
A Complexity-Effective Simultaneous Multithreading Architecture
ICPP '05 Proceedings of the 2005 International Conference on Parallel Processing
Learning-Based SMT Processor Resource Distribution via Hill-Climbing
Proceedings of the 33rd annual international symposium on Computer Architecture
Adaptive reorder buffers for SMT processors
Proceedings of the 15th international conference on Parallel architectures and compilation techniques
Partitioning Multi-Threaded Processors with a Large Number of Threads
ISPASS '05 Proceedings of the IEEE International Symposium on Performance Analysis of Systems and Software, 2005
APPT'05 Proceedings of the 6th international conference on Advanced Parallel Processing Technologies
Hi-index | 0.00 |
An important design issue of SMT processors is to find proper sharing strategies of resources among threads. This paper proposes a ROB sharing strategy, called paired ROB , that considers the fact that task parallelism is not always available to fully utilize resources of multithreaded processors. To this aim, an evaluation methodology is proposed and used for the experiments, which analyzes performance under different degrees of parallelism. Results show that paired ROBs are a cost-effective strategy that provides better performance than private ROBs for low task parallelism, whereas it incurs slight performance losses for high task parallelism.