Importance of single-core performance in the multicore era

Authors:
Toshinori Sato;Hideki Mori;Rikiya Yano;Takanori Hayashida
Affiliations:
Fukuoka University, Nanakuma, Jonan-ku, Fukuoka, Japan;Fukuoka University, Nanakuma, Jonan-ku, Fukuoka, Japan;Fukuoka University, Nanakuma, Jonan-ku, Fukuoka, Japan;Fukuoka University, Nanakuma, Jonan-ku, Fukuoka, Japan
Venue:
ACSC '12 Proceedings of the Thirty-fifth Australasian Computer Science Conference - Volume 122
Year:
2012

Citing 11
Cited 0

Power and energy reduction via pipeline balancing

ISCA '01 Proceedings of the 28th annual international symposium on Computer architecture
SimpleScalar: An Infrastructure for Computer System Modeling

Computer
Mitigating Amdahl's Law through EPI Throttling

Proceedings of the 32nd annual international symposium on Computer Architecture
The Impact of Performance Asymmetry in Emerging Multicore Architectures

Proceedings of the 32nd annual international symposium on Computer Architecture
Heterogeneous Chip Multiprocessors

Computer
Performance, Power Efficiency and Scalability of Asymmetric Cluster Chip Multiprocessors

IEEE Computer Architecture Letters
Core fusion: accommodating software diversity in chip multiprocessors

Proceedings of the 34th annual international symposium on Computer architecture
Thousand core chips: a technology perspective

Proceedings of the 44th annual Design Automation Conference
Dependability, power, and performance trade-off on a multicore processor

Proceedings of the 2008 Asia and South Pacific Design Automation Conference
Amdahl's Law in the Multicore Era

Computer
Investigation of Factors Impacting Thread-Level Parallelism from Desktop, Multimedia and HPC Applications

FCST '09 Proceedings of the 2009 Fourth International Conference on Frontier of Computer Science and Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper first investigates what the best multicore configuration will be in the future, when the number of usable transistors further increases. Comparing five multicore models: single-core, many-core, heterogeneous multicore, scalable homogeneous multicore, and dynamically configurable multicore, surprisingly unveils that single-core performance is a key to improve multicore performance. Based on the findings, this paper secondly proposes a technique to improve single-core performance. It is based on Intel's Turbo Boost technology. From the detailed simulations, it is found that the technique achieves single-core performance improvement.