A disciplined approach to the development of platform architectures

David I. August; K. Keutzer; Sharad Malik; A. R. Newton

doi:10.1016/S0026-2692(02)00069-1

A disciplined approach to the development of platform architectures

David I. August, K. Keutzer, Sharad Malik, A. R. Newton

Research output: Contribution to journal › Article › peer-review

8 Scopus citations

Abstract

Silicon capability has enabled the embedding of an entire system on a single silicon die. These devices are known as systems-on-a-chip. Currently, the design of these devices is undisciplined, expensive, and risky. One way of amortizing the cost and ameliorating this design risk is to make a single integrated circuit serve multiple applications, and the natural way of enabling this is through end-user programmability. The aim of the MESCAL project, which is the subject of this paper, is to introduce a disciplined approach to producing reusable architectural platforms that can be easily programmed to meet a variety of applications. (MESCAL stands for Modern Embedded Systems, Compilers, Architectures, and Languages.)

Original language	English (US)
Pages (from-to)	881-890
Number of pages	10
Journal	Microelectronics Journal
Volume	33
Issue number	11
DOIs	https://doi.org/10.1016/S0026-2692(02)00069-1
State	Published - Nov 2002

All Science Journal Classification (ASJC) codes

Electronic, Optical and Magnetic Materials
Atomic and Molecular Physics, and Optics
Condensed Matter Physics
Surfaces, Coatings and Films
Electrical and Electronic Engineering

Keywords

Application specific instruction set processors (ASIPs)
Mescal project
Platform-architecture development
Retargetable compilation

Access to Document

10.1016/S0026-2692(02)00069-1

Cite this

@article{51c15c5afe03438e96d244f0878c9f5f,

title = "A disciplined approach to the development of platform architectures",

abstract = "Silicon capability has enabled the embedding of an entire system on a single silicon die. These devices are known as systems-on-a-chip. Currently, the design of these devices is undisciplined, expensive, and risky. One way of amortizing the cost and ameliorating this design risk is to make a single integrated circuit serve multiple applications, and the natural way of enabling this is through end-user programmability. The aim of the MESCAL project, which is the subject of this paper, is to introduce a disciplined approach to producing reusable architectural platforms that can be easily programmed to meet a variety of applications. (MESCAL stands for Modern Embedded Systems, Compilers, Architectures, and Languages.)",

keywords = "Application specific instruction set processors (ASIPs), Mescal project, Platform-architecture development, Retargetable compilation",

author = "August, \{David I.\} and K. Keutzer and Sharad Malik and Newton, \{A. R.\}",

year = "2002",

month = nov,

doi = "10.1016/S0026-2692(02)00069-1",

language = "English (US)",

volume = "33",

pages = "881--890",

journal = "Microelectronics Journal",

issn = "0026-2692",

publisher = "Elsevier Ltd",

number = "11",

}

TY - JOUR

T1 - A disciplined approach to the development of platform architectures

AU - August, David I.

AU - Keutzer, K.

AU - Malik, Sharad

AU - Newton, A. R.

PY - 2002/11

Y1 - 2002/11

N2 - Silicon capability has enabled the embedding of an entire system on a single silicon die. These devices are known as systems-on-a-chip. Currently, the design of these devices is undisciplined, expensive, and risky. One way of amortizing the cost and ameliorating this design risk is to make a single integrated circuit serve multiple applications, and the natural way of enabling this is through end-user programmability. The aim of the MESCAL project, which is the subject of this paper, is to introduce a disciplined approach to producing reusable architectural platforms that can be easily programmed to meet a variety of applications. (MESCAL stands for Modern Embedded Systems, Compilers, Architectures, and Languages.)

AB - Silicon capability has enabled the embedding of an entire system on a single silicon die. These devices are known as systems-on-a-chip. Currently, the design of these devices is undisciplined, expensive, and risky. One way of amortizing the cost and ameliorating this design risk is to make a single integrated circuit serve multiple applications, and the natural way of enabling this is through end-user programmability. The aim of the MESCAL project, which is the subject of this paper, is to introduce a disciplined approach to producing reusable architectural platforms that can be easily programmed to meet a variety of applications. (MESCAL stands for Modern Embedded Systems, Compilers, Architectures, and Languages.)

KW - Application specific instruction set processors (ASIPs)

KW - Mescal project

KW - Platform-architecture development

KW - Retargetable compilation

UR - https://www.scopus.com/pages/publications/0036856553

UR - https://www.scopus.com/inward/citedby.url?scp=0036856553&partnerID=8YFLogxK

U2 - 10.1016/S0026-2692(02)00069-1

DO - 10.1016/S0026-2692(02)00069-1

M3 - Article

AN - SCOPUS:0036856553

SN - 0026-2692

VL - 33

SP - 881

EP - 890

JO - Microelectronics Journal

JF - Microelectronics Journal

IS - 11

ER -

A disciplined approach to the development of platform architectures

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this