GTX: the MARCO GSRC technology extrapolation system

Andrew E. Caldwell, Yu Cao, Andrew B. Kahng, Farinaz Koushanfar, Hua Lu, Igor L. Markov, Michael Oliver, Dirk Stroobandt, Dennis Sylvester

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

30 Citations (Scopus)

Abstract

Technology extrapolation - the calibration and prediction of achievable design in future technology generations - drives the evolution of VLSI system architectures, design methodologies, and design tools. This paper describes initial experiences with development and use of GTX, the MARCO GSRC Technology Extrapolation system. GTX provides a robust, portable framework for interactive specification and comparison of modeling choices, e.g., for predicting system cycle time, die size and power dissipation. We use GTX to reveal surprising levels of uncertainty (modeling and parameter sensitivity) in widely-cited cycle-time models that drive recent roadmaps. We also describe new SOI and bulk device models that have been developed for GTX, as well as studies of power dissipation and delay uncertainty under various implementation assumptions for global interconnects.

Original languageEnglish (US)
Title of host publicationProceedings - Design Automation Conference
PublisherIEEE
Pages693-698
Number of pages6
StatePublished - 2000
Externally publishedYes
EventDAC 2000: 37th Design Automation Conference - Los Angeles, CA, USA
Duration: Jun 5 2000Jun 9 2000

Other

OtherDAC 2000: 37th Design Automation Conference
CityLos Angeles, CA, USA
Period6/5/006/9/00

Fingerprint

Extrapolation
Computer systems
Energy dissipation
Calibration
Specifications
Uncertainty

ASJC Scopus subject areas

  • Hardware and Architecture
  • Control and Systems Engineering

Cite this

Caldwell, A. E., Cao, Y., Kahng, A. B., Koushanfar, F., Lu, H., Markov, I. L., ... Sylvester, D. (2000). GTX: the MARCO GSRC technology extrapolation system. In Proceedings - Design Automation Conference (pp. 693-698). IEEE.

GTX : the MARCO GSRC technology extrapolation system. / Caldwell, Andrew E.; Cao, Yu; Kahng, Andrew B.; Koushanfar, Farinaz; Lu, Hua; Markov, Igor L.; Oliver, Michael; Stroobandt, Dirk; Sylvester, Dennis.

Proceedings - Design Automation Conference. IEEE, 2000. p. 693-698.

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

Caldwell, AE, Cao, Y, Kahng, AB, Koushanfar, F, Lu, H, Markov, IL, Oliver, M, Stroobandt, D & Sylvester, D 2000, GTX: the MARCO GSRC technology extrapolation system. in Proceedings - Design Automation Conference. IEEE, pp. 693-698, DAC 2000: 37th Design Automation Conference, Los Angeles, CA, USA, 6/5/00.
Caldwell AE, Cao Y, Kahng AB, Koushanfar F, Lu H, Markov IL et al. GTX: the MARCO GSRC technology extrapolation system. In Proceedings - Design Automation Conference. IEEE. 2000. p. 693-698
Caldwell, Andrew E. ; Cao, Yu ; Kahng, Andrew B. ; Koushanfar, Farinaz ; Lu, Hua ; Markov, Igor L. ; Oliver, Michael ; Stroobandt, Dirk ; Sylvester, Dennis. / GTX : the MARCO GSRC technology extrapolation system. Proceedings - Design Automation Conference. IEEE, 2000. pp. 693-698
@inproceedings{40f36481db5c4f94955eac9c0efefe23,
title = "GTX: the MARCO GSRC technology extrapolation system",
abstract = "Technology extrapolation - the calibration and prediction of achievable design in future technology generations - drives the evolution of VLSI system architectures, design methodologies, and design tools. This paper describes initial experiences with development and use of GTX, the MARCO GSRC Technology Extrapolation system. GTX provides a robust, portable framework for interactive specification and comparison of modeling choices, e.g., for predicting system cycle time, die size and power dissipation. We use GTX to reveal surprising levels of uncertainty (modeling and parameter sensitivity) in widely-cited cycle-time models that drive recent roadmaps. We also describe new SOI and bulk device models that have been developed for GTX, as well as studies of power dissipation and delay uncertainty under various implementation assumptions for global interconnects.",
author = "Caldwell, {Andrew E.} and Yu Cao and Kahng, {Andrew B.} and Farinaz Koushanfar and Hua Lu and Markov, {Igor L.} and Michael Oliver and Dirk Stroobandt and Dennis Sylvester",
year = "2000",
language = "English (US)",
pages = "693--698",
booktitle = "Proceedings - Design Automation Conference",
publisher = "IEEE",

}

TY - GEN

T1 - GTX

T2 - the MARCO GSRC technology extrapolation system

AU - Caldwell, Andrew E.

AU - Cao, Yu

AU - Kahng, Andrew B.

AU - Koushanfar, Farinaz

AU - Lu, Hua

AU - Markov, Igor L.

AU - Oliver, Michael

AU - Stroobandt, Dirk

AU - Sylvester, Dennis

PY - 2000

Y1 - 2000

N2 - Technology extrapolation - the calibration and prediction of achievable design in future technology generations - drives the evolution of VLSI system architectures, design methodologies, and design tools. This paper describes initial experiences with development and use of GTX, the MARCO GSRC Technology Extrapolation system. GTX provides a robust, portable framework for interactive specification and comparison of modeling choices, e.g., for predicting system cycle time, die size and power dissipation. We use GTX to reveal surprising levels of uncertainty (modeling and parameter sensitivity) in widely-cited cycle-time models that drive recent roadmaps. We also describe new SOI and bulk device models that have been developed for GTX, as well as studies of power dissipation and delay uncertainty under various implementation assumptions for global interconnects.

AB - Technology extrapolation - the calibration and prediction of achievable design in future technology generations - drives the evolution of VLSI system architectures, design methodologies, and design tools. This paper describes initial experiences with development and use of GTX, the MARCO GSRC Technology Extrapolation system. GTX provides a robust, portable framework for interactive specification and comparison of modeling choices, e.g., for predicting system cycle time, die size and power dissipation. We use GTX to reveal surprising levels of uncertainty (modeling and parameter sensitivity) in widely-cited cycle-time models that drive recent roadmaps. We also describe new SOI and bulk device models that have been developed for GTX, as well as studies of power dissipation and delay uncertainty under various implementation assumptions for global interconnects.

UR - http://www.scopus.com/inward/record.url?scp=0033720599&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033720599&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0033720599

SP - 693

EP - 698

BT - Proceedings - Design Automation Conference

PB - IEEE

ER -