Reducing memory requirements through task recomputation in embedded multi-CPU systems

H. Koc, S. Tosun, O. Ozturk, Mahmut Kandemir

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

6 Citations (Scopus)

Abstract

As embedded applications are processing increasingly larger data sets, keeping their memory space consumptions under control is becoming a very pressing issue. Observing this, several prior efforts have considered memory space reduction techniques (in both hardware and software) based on data compression and lifetime-based memory recycling (garbage collection). In this work, we propose and evaluate an alternate approach to memory space saving in multi-CPU embedded systems such as chip multiprocessors. The unique characteristic of our approach is that it recomputes the results of select tasks in a given task graph (which represents the application), instead of storing these results in memory and accessing them from there as needed.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006
Pages448-449
Number of pages2
Volume2006
DOIs
StatePublished - Oct 9 2006
EventIEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006 - Klarlsruhe, Germany
Duration: Mar 2 2006Mar 3 2006

Other

OtherIEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006
CountryGermany
CityKlarlsruhe
Period3/2/063/3/06

Fingerprint

Program processors
Data storage equipment
Data compression
Embedded systems
Computer hardware
Recycling
Processing

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Koc, H., Tosun, S., Ozturk, O., & Kandemir, M. (2006). Reducing memory requirements through task recomputation in embedded multi-CPU systems. In Proceedings - IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006 (Vol. 2006, pp. 448-449). [1602490] https://doi.org/10.1109/ISVLSI.2006.77
Koc, H. ; Tosun, S. ; Ozturk, O. ; Kandemir, Mahmut. / Reducing memory requirements through task recomputation in embedded multi-CPU systems. Proceedings - IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006. Vol. 2006 2006. pp. 448-449
@inproceedings{3c2d6db0e0d44b81976f64cd461c4100,
title = "Reducing memory requirements through task recomputation in embedded multi-CPU systems",
abstract = "As embedded applications are processing increasingly larger data sets, keeping their memory space consumptions under control is becoming a very pressing issue. Observing this, several prior efforts have considered memory space reduction techniques (in both hardware and software) based on data compression and lifetime-based memory recycling (garbage collection). In this work, we propose and evaluate an alternate approach to memory space saving in multi-CPU embedded systems such as chip multiprocessors. The unique characteristic of our approach is that it recomputes the results of select tasks in a given task graph (which represents the application), instead of storing these results in memory and accessing them from there as needed.",
author = "H. Koc and S. Tosun and O. Ozturk and Mahmut Kandemir",
year = "2006",
month = "10",
day = "9",
doi = "10.1109/ISVLSI.2006.77",
language = "English (US)",
isbn = "0769525334",
volume = "2006",
pages = "448--449",
booktitle = "Proceedings - IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006",

}

Koc, H, Tosun, S, Ozturk, O & Kandemir, M 2006, Reducing memory requirements through task recomputation in embedded multi-CPU systems. in Proceedings - IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006. vol. 2006, 1602490, pp. 448-449, IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006, Klarlsruhe, Germany, 3/2/06. https://doi.org/10.1109/ISVLSI.2006.77

Reducing memory requirements through task recomputation in embedded multi-CPU systems. / Koc, H.; Tosun, S.; Ozturk, O.; Kandemir, Mahmut.

Proceedings - IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006. Vol. 2006 2006. p. 448-449 1602490.

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

TY - GEN

T1 - Reducing memory requirements through task recomputation in embedded multi-CPU systems

AU - Koc, H.

AU - Tosun, S.

AU - Ozturk, O.

AU - Kandemir, Mahmut

PY - 2006/10/9

Y1 - 2006/10/9

N2 - As embedded applications are processing increasingly larger data sets, keeping their memory space consumptions under control is becoming a very pressing issue. Observing this, several prior efforts have considered memory space reduction techniques (in both hardware and software) based on data compression and lifetime-based memory recycling (garbage collection). In this work, we propose and evaluate an alternate approach to memory space saving in multi-CPU embedded systems such as chip multiprocessors. The unique characteristic of our approach is that it recomputes the results of select tasks in a given task graph (which represents the application), instead of storing these results in memory and accessing them from there as needed.

AB - As embedded applications are processing increasingly larger data sets, keeping their memory space consumptions under control is becoming a very pressing issue. Observing this, several prior efforts have considered memory space reduction techniques (in both hardware and software) based on data compression and lifetime-based memory recycling (garbage collection). In this work, we propose and evaluate an alternate approach to memory space saving in multi-CPU embedded systems such as chip multiprocessors. The unique characteristic of our approach is that it recomputes the results of select tasks in a given task graph (which represents the application), instead of storing these results in memory and accessing them from there as needed.

UR - http://www.scopus.com/inward/record.url?scp=33749365593&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33749365593&partnerID=8YFLogxK

U2 - 10.1109/ISVLSI.2006.77

DO - 10.1109/ISVLSI.2006.77

M3 - Conference contribution

AN - SCOPUS:33749365593

SN - 0769525334

SN - 9780769525334

VL - 2006

SP - 448

EP - 449

BT - Proceedings - IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006

ER -

Koc H, Tosun S, Ozturk O, Kandemir M. Reducing memory requirements through task recomputation in embedded multi-CPU systems. In Proceedings - IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures 2006. Vol. 2006. 2006. p. 448-449. 1602490 https://doi.org/10.1109/ISVLSI.2006.77