Automatic data migration for reducing energy consumption in multi-bank memory systems

V. De La Luz, Mahmut Kandemir, I. Kolcu

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

55 Citations (Scopus)

Abstract

An architectural solution to reducing memory energy consumption is to adopt a multi-bank memory system instead of a monolithic (single-bank) memory system. Some recent multi-bank memory architectures help reduce memory energy by allowing an unused bank to be placed into a low-power operating mode. This paper describes an automatic data migration strategy which dynamically places the arrays with temporal affinity into the same set of banks. This strategy increases the number of banks which can be put into low-power modes and allows the use of more aggressive energysaving modes. Experiments using several array-dominated applications show the usefulness of data migration and indicate that large energy savings can be achieved with low overhead.

Original languageEnglish (US)
Title of host publicationProceedings of the 39th Annual Design Automation Conference, DAC'02
Pages213-218
Number of pages6
StatePublished - Aug 31 2002
Event39th Annual Design Automation Conference, DAC'02 - New Orleans, LA, United States
Duration: Jun 10 2002Jun 14 2002

Other

Other39th Annual Design Automation Conference, DAC'02
CountryUnited States
CityNew Orleans, LA
Period6/10/026/14/02

Fingerprint

Computer systems
Energy utilization
Data storage equipment
Memory architecture
Energy conservation
Experiments

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Control and Systems Engineering

Cite this

De La Luz, V., Kandemir, M., & Kolcu, I. (2002). Automatic data migration for reducing energy consumption in multi-bank memory systems. In Proceedings of the 39th Annual Design Automation Conference, DAC'02 (pp. 213-218)
De La Luz, V. ; Kandemir, Mahmut ; Kolcu, I. / Automatic data migration for reducing energy consumption in multi-bank memory systems. Proceedings of the 39th Annual Design Automation Conference, DAC'02. 2002. pp. 213-218
@inproceedings{85ffd9b07bdf4ed2a254cf302d41a534,
title = "Automatic data migration for reducing energy consumption in multi-bank memory systems",
abstract = "An architectural solution to reducing memory energy consumption is to adopt a multi-bank memory system instead of a monolithic (single-bank) memory system. Some recent multi-bank memory architectures help reduce memory energy by allowing an unused bank to be placed into a low-power operating mode. This paper describes an automatic data migration strategy which dynamically places the arrays with temporal affinity into the same set of banks. This strategy increases the number of banks which can be put into low-power modes and allows the use of more aggressive energysaving modes. Experiments using several array-dominated applications show the usefulness of data migration and indicate that large energy savings can be achieved with low overhead.",
author = "{De La Luz}, V. and Mahmut Kandemir and I. Kolcu",
year = "2002",
month = "8",
day = "31",
language = "English (US)",
isbn = "1581134614",
pages = "213--218",
booktitle = "Proceedings of the 39th Annual Design Automation Conference, DAC'02",

}

De La Luz, V, Kandemir, M & Kolcu, I 2002, Automatic data migration for reducing energy consumption in multi-bank memory systems. in Proceedings of the 39th Annual Design Automation Conference, DAC'02. pp. 213-218, 39th Annual Design Automation Conference, DAC'02, New Orleans, LA, United States, 6/10/02.

Automatic data migration for reducing energy consumption in multi-bank memory systems. / De La Luz, V.; Kandemir, Mahmut; Kolcu, I.

Proceedings of the 39th Annual Design Automation Conference, DAC'02. 2002. p. 213-218.

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

TY - GEN

T1 - Automatic data migration for reducing energy consumption in multi-bank memory systems

AU - De La Luz, V.

AU - Kandemir, Mahmut

AU - Kolcu, I.

PY - 2002/8/31

Y1 - 2002/8/31

N2 - An architectural solution to reducing memory energy consumption is to adopt a multi-bank memory system instead of a monolithic (single-bank) memory system. Some recent multi-bank memory architectures help reduce memory energy by allowing an unused bank to be placed into a low-power operating mode. This paper describes an automatic data migration strategy which dynamically places the arrays with temporal affinity into the same set of banks. This strategy increases the number of banks which can be put into low-power modes and allows the use of more aggressive energysaving modes. Experiments using several array-dominated applications show the usefulness of data migration and indicate that large energy savings can be achieved with low overhead.

AB - An architectural solution to reducing memory energy consumption is to adopt a multi-bank memory system instead of a monolithic (single-bank) memory system. Some recent multi-bank memory architectures help reduce memory energy by allowing an unused bank to be placed into a low-power operating mode. This paper describes an automatic data migration strategy which dynamically places the arrays with temporal affinity into the same set of banks. This strategy increases the number of banks which can be put into low-power modes and allows the use of more aggressive energysaving modes. Experiments using several array-dominated applications show the usefulness of data migration and indicate that large energy savings can be achieved with low overhead.

UR - http://www.scopus.com/inward/record.url?scp=0036059441&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036059441&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1581134614

SP - 213

EP - 218

BT - Proceedings of the 39th Annual Design Automation Conference, DAC'02

ER -

De La Luz V, Kandemir M, Kolcu I. Automatic data migration for reducing energy consumption in multi-bank memory systems. In Proceedings of the 39th Annual Design Automation Conference, DAC'02. 2002. p. 213-218