Speculative High Performance Computation on Heterogeneous Multi-Core

Cong Liu; Wen Wang; Zhi Ying Wang

doi:10.4028/www.scientific.net/AMR.1049-1050.2126

Paper Titles

Approach to the Management of Warfare Simulation Scenarios Based on Ontology and CBR
p.2105

Comprehensive Evaluation on Software Scheme of Sport E-Government Platform
p.2114

Research on the Present Development Status and Path Construction of Domestic E-Government
p.2118

Research on Cloud Computing Environment Energy Saving Optimization Method
p.2122

Speculative High Performance Computation on Heterogeneous Multi-Core
p.2126

Protection Approach against SRLG Failure for Survivable Long-Reach Passive Optical Network in Smart Grid
p.2131

Implementation of IEEE 802.11s in OPNET
p.2135

Framework and Design of the Enterprise Integration Based on SaaS Platform
p.2139

An Improved Localization Algorithm in Wireless Sensor Networks
p.2144

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 1049-1050Speculative High Performance Computation on...

Speculative High Performance Computation on Heterogeneous Multi-Core

Abstract:

Thread level speculation has been proposed and researched to parallelize traditional sequential applications on homogeneous multi-core architecture. In this paper, a heterogeneous multi-core hardware simulation system is present, which provides with TLS execution mechanism. With a novel TLS programming model and a number of new speculative tuning techniques, benchmark Gzip is parallelized from-3% to 195% on a four-core processor, and the speedup of the test benchmarks are 30%, 43% and 156%, respectively with arbitrary, hotspot and insight speculation.

You might also be interested in these eBooks

Modern Technologies in Materials, Mechanics and Intelligent Systems

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 1049-1050)

Pages:

2126-2130

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.1049-1050.2126

Citation:

Cite this paper

Online since:

October 2014

Authors:

Cong Liu*, Wen Wang, Zhi Ying Wang

Keywords:

Heterogeneous Multi-Core, High Performance Computing, Speculative Threads

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] L. Hammond, B.A. Hubbert, et al: The Stanford Hydra CMP , IEEE Micro(2000), 71.

Google Scholar

[2] J. Renau, J. Tuck, et al: Tasking with Out-of-Order Spawn in TLS Chip Multiprocessors: Microarchitecture and Compilation, ICS(2005).

DOI: 10.1145/1088149.1088173

Google Scholar

[3] SESC simulator, http: /sesc. sourceforge. net.

Google Scholar

[4] W. Liu, J. Tuck, et al: POSH: a TLS compiler that exploits program structure, PPoPP(2006).

Google Scholar

[5] J. W. Davidson, S. Jinturkar: Improving instruction-level parallelism by loop unrolling and dynamic memory disambiguation, Micro(1995).

DOI: 10.1109/micro.1995.476820

Google Scholar

[6] K. Olukotun, L. Hammond, J. Laudon: Chip Multiprocessor Architecture Techniques to Improve Throughput and Latency, San Francisco: M&C Publisher, (2008).

DOI: 10.1007/978-3-031-01720-9

Google Scholar

[7] C. Ding, X. Shen, et al: Software behavior oriented parallelization, PLDI(2007).

Google Scholar

[8] V. Krishnan and J. Torrellas: A Chip-Multiprocessor Architecture with Speculative Multithreading, IEEE Trans. Comput(1999), 866.

DOI: 10.1109/12.795218

Google Scholar

[9] http: /www. gzip. org.

Google Scholar

[10] J. Gilchrist: Parallel data compression with Bzip2, PDCS(2004).

Google Scholar

[11] R. Kumar, K.I. Farks, et al: Single-ISA heterogeneous multi-core architectures: The potential for processor power reduction, MICRO(2003).

Google Scholar

[12] R. Kumar, D. M. Tullsen, et al: Single-ISA heterogeneous multi-core architectures for multithreaded workload performance, ISCA(2004).

DOI: 10.1109/isca.2004.1310764

Google Scholar

[13] Y. Luo, W. Hsu: The Design and Implementation of Heterogeneous Multicore Systems for Energy-efficient Speculative Thread Execution, ACM Trans. on Architecture and Code Optimization(2013).

DOI: 10.1145/2541228.2541233

Google Scholar