Adapt or Become Extinct! : The Case for a Unified Framework for Deployment-Time Optimization

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2011

Standard

Adapt or Become Extinct! : The Case for a Unified Framework for Deployment-Time Optimization. / Goumas, Georgios; McKee, Sally A.; Själander, Magnus; Gross, Thomas R.; Karlsson, Sven; Probst, Christian W.; Zhang, Lixin.

EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era. University of Strathclyde, 2011.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2011

Harvard

Goumas, G, McKee, SA, Själander, M, Gross, TR, Karlsson, S, Probst, CW & Zhang, L 2011, 'Adapt or Become Extinct!: The Case for a Unified Framework for Deployment-Time Optimization'. in EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era. University of Strathclyde., 10.1145/2000417.2000422

APA

Goumas, G., McKee, S. A., Själander, M., Gross, T. R., Karlsson, S., Probst, C. W., & Zhang, L. (2011). Adapt or Become Extinct!: The Case for a Unified Framework for Deployment-Time Optimization. In EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era. University of Strathclyde. 10.1145/2000417.2000422

CBE

Goumas G, McKee SA, Själander M, Gross TR, Karlsson S, Probst CW, Zhang L. 2011. Adapt or Become Extinct!: The Case for a Unified Framework for Deployment-Time Optimization. In EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era. University of Strathclyde. Available from: 10.1145/2000417.2000422

MLA

Goumas, Georgios et al. "Adapt or Become Extinct!: The Case for a Unified Framework for Deployment-Time Optimization". EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era. University of Strathclyde. 2011. Available: 10.1145/2000417.2000422

Vancouver

Goumas G, McKee SA, Själander M, Gross TR, Karlsson S, Probst CW et al. Adapt or Become Extinct!: The Case for a Unified Framework for Deployment-Time Optimization. In EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era. University of Strathclyde. 2011. Available from: 10.1145/2000417.2000422

Author

Goumas, Georgios; McKee, Sally A.; Själander, Magnus; Gross, Thomas R.; Karlsson, Sven; Probst, Christian W.; Zhang, Lixin / Adapt or Become Extinct! : The Case for a Unified Framework for Deployment-Time Optimization.

EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era. University of Strathclyde, 2011.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2011

Bibtex

@inbook{7346eade452b4a5ea00693fcd963718e,
title = "Adapt or Become Extinct!",
publisher = "University of Strathclyde",
author = "Georgios Goumas and McKee, {Sally A.} and Magnus Själander and Gross, {Thomas R.} and Sven Karlsson and Probst, {Christian W.} and Lixin Zhang",
year = "2011",
doi = "10.1145/2000417.2000422",
isbn = "978-1-4503-0708-6",
booktitle = "EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era",

}

RIS

TY - GEN

T1 - Adapt or Become Extinct!

T2 - EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era

A1 - Goumas,Georgios

A1 - McKee,Sally A.

A1 - Själander,Magnus

A1 - Gross,Thomas R.

A1 - Karlsson,Sven

A1 - Probst,Christian W.

A1 - Zhang,Lixin

AU - Goumas,Georgios

AU - McKee,Sally A.

AU - Själander,Magnus

AU - Gross,Thomas R.

AU - Karlsson,Sven

AU - Probst,Christian W.

AU - Zhang,Lixin

PB - University of Strathclyde

PY - 2011

Y1 - 2011

N2 - The High-Performance Computing ecosystem consists of a large variety of execution platforms that demonstrate a wide diversity in hardware characteristics such as CPU architecture, memory organization, interconnection network, accelerators, etc. This environment also presents a number of hard boundaries (walls) for applications which limit software development (parallel programming wall), performance (memory wall, communication wall) and viability (power wall). The only way to survive in such a demanding environment is by adaptation. In this paper we discuss how dynamic information collected during the execution of an application can be utilized to adapt the execution context and may lead to performance gains beyond those provided by static information and compile-time adaptation. We consider specialization based on dynamic information like user input, architectural characteristics such as the memory hierarchy organization, and the execution prole of the application as obtained from the execution platform's performance monitoring units. One of the challenges of future execution platforms is to allow the seamless integration of these various kinds of information with information obtained from static analysis (either during ahead-of-time or just-in-time) compilation. We extend the notion of information-driven adaptation and outline the architecture of an infrastructure designed to enable information ow and adaptation throughout the life-cycle of an application.

AB - The High-Performance Computing ecosystem consists of a large variety of execution platforms that demonstrate a wide diversity in hardware characteristics such as CPU architecture, memory organization, interconnection network, accelerators, etc. This environment also presents a number of hard boundaries (walls) for applications which limit software development (parallel programming wall), performance (memory wall, communication wall) and viability (power wall). The only way to survive in such a demanding environment is by adaptation. In this paper we discuss how dynamic information collected during the execution of an application can be utilized to adapt the execution context and may lead to performance gains beyond those provided by static information and compile-time adaptation. We consider specialization based on dynamic information like user input, architectural characteristics such as the memory hierarchy organization, and the execution prole of the application as obtained from the execution platform's performance monitoring units. One of the challenges of future execution platforms is to allow the seamless integration of these various kinds of information with information obtained from static analysis (either during ahead-of-time or just-in-time) compilation. We extend the notion of information-driven adaptation and outline the architecture of an infrastructure designed to enable information ow and adaptation throughout the life-cycle of an application.

UR - http://exadapt.org/2011/

U2 - 10.1145/2000417.2000422

DO - 10.1145/2000417.2000422

SN - 978-1-4503-0708-6

BT - EXADAPT '11 Proceedings of the 1st International Workshop on Adaptive Self-Tuning Computing Systems for the Exaflop Era

ER -