Systematic prototyping of superscalar computer architectures

Thomas M. Conte, Wen-Mei W Hwu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

There are a staggering number of design decisions that must be made to design a computer architecture. These are typically resolved using simulation techniques that use inputs composed of a set of traces of instructions taken from executing programs. These traces are collectively called the workload. The workload dictates the architectural-level design. The bottleneck in this architectural design process is the the detailed simulation. One solution that has is to limit the size of the workload used. Although this improves the speed of the design process, it limits the quality of the final design. Another approach is to use published results, yet little or no coherence exists between the workloads that one architecture researcher uses and those used by another. This paper argues that the correct solution to the computer architecture design process is a prototyping approach. In this approach, the initial design is selected using the workload itself. This design is an architectural prototype, specifying the high-level design decisions that are difficult to acquire via detailed simulation. These design decisions include what the mix of function units in the execution stage of the processor is and what the dimensions of the caches in the memory subsystem should be. This paper selects architectural prototypes by trading off accuracy in hardware simulation for an increase in usable workload size.

Original languageEnglish (US)
Title of host publicationProceedings - 3rd International Workshop on Rapid System Prototyping
Subtitle of host publicationShortening the Path from Specification to Prototype, RSP 1992
PublisherIEEE Computer Society
Pages161-170
Number of pages10
ISBN (Electronic)0818635207
DOIs
StatePublished - Jan 1 1992
Event3rd International Workshop on Rapid System Prototyping, RSP 1992 - Durham, United States
Duration: Jun 23 1992Jun 25 1992

Publication series

NameProceedings of the International Workshop on Rapid System Prototyping
Volume1992-January
ISSN (Print)1074-6005

Other

Other3rd International Workshop on Rapid System Prototyping, RSP 1992
CountryUnited States
CityDurham
Period6/23/926/25/92

Fingerprint

Computer architecture
Architectural design
Hardware
Data storage equipment

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Conte, T. M., & Hwu, W-M. W. (1992). Systematic prototyping of superscalar computer architectures. In Proceedings - 3rd International Workshop on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 1992 (pp. 161-170). [243910] (Proceedings of the International Workshop on Rapid System Prototyping; Vol. 1992-January). IEEE Computer Society. https://doi.org/10.1109/IWRSP.1992.243910

Systematic prototyping of superscalar computer architectures. / Conte, Thomas M.; Hwu, Wen-Mei W.

Proceedings - 3rd International Workshop on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 1992. IEEE Computer Society, 1992. p. 161-170 243910 (Proceedings of the International Workshop on Rapid System Prototyping; Vol. 1992-January).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Conte, TM & Hwu, W-MW 1992, Systematic prototyping of superscalar computer architectures. in Proceedings - 3rd International Workshop on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 1992., 243910, Proceedings of the International Workshop on Rapid System Prototyping, vol. 1992-January, IEEE Computer Society, pp. 161-170, 3rd International Workshop on Rapid System Prototyping, RSP 1992, Durham, United States, 6/23/92. https://doi.org/10.1109/IWRSP.1992.243910
Conte TM, Hwu W-MW. Systematic prototyping of superscalar computer architectures. In Proceedings - 3rd International Workshop on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 1992. IEEE Computer Society. 1992. p. 161-170. 243910. (Proceedings of the International Workshop on Rapid System Prototyping). https://doi.org/10.1109/IWRSP.1992.243910
Conte, Thomas M. ; Hwu, Wen-Mei W. / Systematic prototyping of superscalar computer architectures. Proceedings - 3rd International Workshop on Rapid System Prototyping: Shortening the Path from Specification to Prototype, RSP 1992. IEEE Computer Society, 1992. pp. 161-170 (Proceedings of the International Workshop on Rapid System Prototyping).
@inproceedings{8498464951664ae2b82cb5cf7b20b7ae,
title = "Systematic prototyping of superscalar computer architectures",
abstract = "There are a staggering number of design decisions that must be made to design a computer architecture. These are typically resolved using simulation techniques that use inputs composed of a set of traces of instructions taken from executing programs. These traces are collectively called the workload. The workload dictates the architectural-level design. The bottleneck in this architectural design process is the the detailed simulation. One solution that has is to limit the size of the workload used. Although this improves the speed of the design process, it limits the quality of the final design. Another approach is to use published results, yet little or no coherence exists between the workloads that one architecture researcher uses and those used by another. This paper argues that the correct solution to the computer architecture design process is a prototyping approach. In this approach, the initial design is selected using the workload itself. This design is an architectural prototype, specifying the high-level design decisions that are difficult to acquire via detailed simulation. These design decisions include what the mix of function units in the execution stage of the processor is and what the dimensions of the caches in the memory subsystem should be. This paper selects architectural prototypes by trading off accuracy in hardware simulation for an increase in usable workload size.",
author = "Conte, {Thomas M.} and Hwu, {Wen-Mei W}",
year = "1992",
month = "1",
day = "1",
doi = "10.1109/IWRSP.1992.243910",
language = "English (US)",
series = "Proceedings of the International Workshop on Rapid System Prototyping",
publisher = "IEEE Computer Society",
pages = "161--170",
booktitle = "Proceedings - 3rd International Workshop on Rapid System Prototyping",

}

TY - GEN

T1 - Systematic prototyping of superscalar computer architectures

AU - Conte, Thomas M.

AU - Hwu, Wen-Mei W

PY - 1992/1/1

Y1 - 1992/1/1

N2 - There are a staggering number of design decisions that must be made to design a computer architecture. These are typically resolved using simulation techniques that use inputs composed of a set of traces of instructions taken from executing programs. These traces are collectively called the workload. The workload dictates the architectural-level design. The bottleneck in this architectural design process is the the detailed simulation. One solution that has is to limit the size of the workload used. Although this improves the speed of the design process, it limits the quality of the final design. Another approach is to use published results, yet little or no coherence exists between the workloads that one architecture researcher uses and those used by another. This paper argues that the correct solution to the computer architecture design process is a prototyping approach. In this approach, the initial design is selected using the workload itself. This design is an architectural prototype, specifying the high-level design decisions that are difficult to acquire via detailed simulation. These design decisions include what the mix of function units in the execution stage of the processor is and what the dimensions of the caches in the memory subsystem should be. This paper selects architectural prototypes by trading off accuracy in hardware simulation for an increase in usable workload size.

AB - There are a staggering number of design decisions that must be made to design a computer architecture. These are typically resolved using simulation techniques that use inputs composed of a set of traces of instructions taken from executing programs. These traces are collectively called the workload. The workload dictates the architectural-level design. The bottleneck in this architectural design process is the the detailed simulation. One solution that has is to limit the size of the workload used. Although this improves the speed of the design process, it limits the quality of the final design. Another approach is to use published results, yet little or no coherence exists between the workloads that one architecture researcher uses and those used by another. This paper argues that the correct solution to the computer architecture design process is a prototyping approach. In this approach, the initial design is selected using the workload itself. This design is an architectural prototype, specifying the high-level design decisions that are difficult to acquire via detailed simulation. These design decisions include what the mix of function units in the execution stage of the processor is and what the dimensions of the caches in the memory subsystem should be. This paper selects architectural prototypes by trading off accuracy in hardware simulation for an increase in usable workload size.

UR - http://www.scopus.com/inward/record.url?scp=84948760905&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948760905&partnerID=8YFLogxK

U2 - 10.1109/IWRSP.1992.243910

DO - 10.1109/IWRSP.1992.243910

M3 - Conference contribution

AN - SCOPUS:84948760905

T3 - Proceedings of the International Workshop on Rapid System Prototyping

SP - 161

EP - 170

BT - Proceedings - 3rd International Workshop on Rapid System Prototyping

PB - IEEE Computer Society

ER -