Embedding a superscalar processor onto a chip multiprocessor

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Chip multiprocessors (CMPs) aim to develop both instruction-level and thread-level parallelisms to boost a system's performance. However, according to previous research results, CMPs outperform superscalar processors only in floating-point applications. Therefore, we have proposed a novel microprocessor, supporting two execution modes, to allow users to manually choose an appropriate mode to execute an application according to the workload characteristics. In the first mode, the processor acts like a conventional CMP. The second mode is derived from aggregating multiple processing elements in the CMP into a wide superscalar. Furthermore, we extend this innovative microarchitecture to support a third execution mode, whereby the processor keeps switching between the first and second modes when executing an application, according to the characteristics of subsequent instructions. As a result, this third mode can use both the advantages of a CMP and of a superscalar to execute an application. According to the performance analysis, our processor can provide an optimum system performance for all benchmark programs, regardless of workload characteristics. Furthermore, our CMP outperforms a conventional CMP, exhibiting a speedup of up to 1.32.

Original languageEnglish
Pages (from-to)147-156
Number of pages10
JournalMicroprocessors and Microsystems
Volume28
Issue number4
DOIs
Publication statusPublished - 2004 May 20

Fingerprint

Microprocessor chips
Processing

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

@article{5153344cff6d467da99812cceee9f639,
title = "Embedding a superscalar processor onto a chip multiprocessor",
abstract = "Chip multiprocessors (CMPs) aim to develop both instruction-level and thread-level parallelisms to boost a system's performance. However, according to previous research results, CMPs outperform superscalar processors only in floating-point applications. Therefore, we have proposed a novel microprocessor, supporting two execution modes, to allow users to manually choose an appropriate mode to execute an application according to the workload characteristics. In the first mode, the processor acts like a conventional CMP. The second mode is derived from aggregating multiple processing elements in the CMP into a wide superscalar. Furthermore, we extend this innovative microarchitecture to support a third execution mode, whereby the processor keeps switching between the first and second modes when executing an application, according to the characteristics of subsequent instructions. As a result, this third mode can use both the advantages of a CMP and of a superscalar to execute an application. According to the performance analysis, our processor can provide an optimum system performance for all benchmark programs, regardless of workload characteristics. Furthermore, our CMP outperforms a conventional CMP, exhibiting a speedup of up to 1.32.",
author = "Chao-Chin Wu",
year = "2004",
month = "5",
day = "20",
doi = "10.1016/j.micpro.2004.02.002",
language = "English",
volume = "28",
pages = "147--156",
journal = "Microprocessors and Microsystems",
issn = "0141-9331",
publisher = "Elsevier",
number = "4",

}

Embedding a superscalar processor onto a chip multiprocessor. / Wu, Chao-Chin.

In: Microprocessors and Microsystems, Vol. 28, No. 4, 20.05.2004, p. 147-156.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Embedding a superscalar processor onto a chip multiprocessor

AU - Wu, Chao-Chin

PY - 2004/5/20

Y1 - 2004/5/20

N2 - Chip multiprocessors (CMPs) aim to develop both instruction-level and thread-level parallelisms to boost a system's performance. However, according to previous research results, CMPs outperform superscalar processors only in floating-point applications. Therefore, we have proposed a novel microprocessor, supporting two execution modes, to allow users to manually choose an appropriate mode to execute an application according to the workload characteristics. In the first mode, the processor acts like a conventional CMP. The second mode is derived from aggregating multiple processing elements in the CMP into a wide superscalar. Furthermore, we extend this innovative microarchitecture to support a third execution mode, whereby the processor keeps switching between the first and second modes when executing an application, according to the characteristics of subsequent instructions. As a result, this third mode can use both the advantages of a CMP and of a superscalar to execute an application. According to the performance analysis, our processor can provide an optimum system performance for all benchmark programs, regardless of workload characteristics. Furthermore, our CMP outperforms a conventional CMP, exhibiting a speedup of up to 1.32.

AB - Chip multiprocessors (CMPs) aim to develop both instruction-level and thread-level parallelisms to boost a system's performance. However, according to previous research results, CMPs outperform superscalar processors only in floating-point applications. Therefore, we have proposed a novel microprocessor, supporting two execution modes, to allow users to manually choose an appropriate mode to execute an application according to the workload characteristics. In the first mode, the processor acts like a conventional CMP. The second mode is derived from aggregating multiple processing elements in the CMP into a wide superscalar. Furthermore, we extend this innovative microarchitecture to support a third execution mode, whereby the processor keeps switching between the first and second modes when executing an application, according to the characteristics of subsequent instructions. As a result, this third mode can use both the advantages of a CMP and of a superscalar to execute an application. According to the performance analysis, our processor can provide an optimum system performance for all benchmark programs, regardless of workload characteristics. Furthermore, our CMP outperforms a conventional CMP, exhibiting a speedup of up to 1.32.

UR - http://www.scopus.com/inward/record.url?scp=2042509000&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2042509000&partnerID=8YFLogxK

U2 - 10.1016/j.micpro.2004.02.002

DO - 10.1016/j.micpro.2004.02.002

M3 - Article

AN - SCOPUS:2042509000

VL - 28

SP - 147

EP - 156

JO - Microprocessors and Microsystems

JF - Microprocessors and Microsystems

SN - 0141-9331

IS - 4

ER -