Dynamic scratch-pad memory management with data pipelining for embedded systems

Yanqin Yang; Meng Wang; Zili Shao; Minyi Guo

doi:10.1109/CSE.2009.295

Dynamic scratch-pad memory management with data pipelining for embedded systems

Yanqin Yang
, Meng Wang
, Zili Shao
, Minyi Guo

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

5 Citations (Scopus)

Abstract

In this paper, we propose an effective data pipelining technique, SPDP (scratch-pad data pipelining), for dynamic scratch-pad memory (SPM) management with DMA (Direct Memory Access). In SPDP, we group multiple iterations of a loop into a block for SPM allocation, and implement a data pipeline by overlapping the execution of CPU instructions and DMA operations. We have implemented our SPDP technique into the IMPACT compiler, and conduct experiments using a set of benchmarks from DSPstone, Mibench and Mediabench on the cycle-accurate VLIW simulator of Trimaran. The experimental results show that our technique achieves significant performance improvement compared with the previous work.

Original language	English
Title of host publication	Proceedings - 12th IEEE International Conference on Computational Science and Engineering, CSE 2009 - 7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009
Pages	358-365
Number of pages	8
Volume	2
DOIs	https://doi.org/10.1109/CSE.2009.295
Publication status	Published - 3 Dec 2009
Event	7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009 - Vancouver, BC, Canada Duration: 29 Aug 2009 → 31 Aug 2009

Conference

Conference	7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009
Country/Territory	Canada
City	Vancouver, BC
Period	29/08/09 → 31/08/09

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Science Applications
Software

More information

10.1109/CSE.2009.295

Cite this

Yang, Y., Wang, M., Shao, Z., & Guo, M. (2009). Dynamic scratch-pad memory management with data pipelining for embedded systems. In Proceedings - 12th IEEE International Conference on Computational Science and Engineering, CSE 2009 - 7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009 (Vol. 2, pp. 358-365). Article 5284058 https://doi.org/10.1109/CSE.2009.295

@inproceedings{cdfcec2fac5f4b9783d44e06ba4214f2,

title = "Dynamic scratch-pad memory management with data pipelining for embedded systems",

abstract = "In this paper, we propose an effective data pipelining technique, SPDP (scratch-pad data pipelining), for dynamic scratch-pad memory (SPM) management with DMA (Direct Memory Access). In SPDP, we group multiple iterations of a loop into a block for SPM allocation, and implement a data pipeline by overlapping the execution of CPU instructions and DMA operations. We have implemented our SPDP technique into the IMPACT compiler, and conduct experiments using a set of benchmarks from DSPstone, Mibench and Mediabench on the cycle-accurate VLIW simulator of Trimaran. The experimental results show that our technique achieves significant performance improvement compared with the previous work.",

author = "Yanqin Yang and Meng Wang and Zili Shao and Minyi Guo",

year = "2009",

month = dec,

day = "3",

doi = "10.1109/CSE.2009.295",

language = "English",

isbn = "9780769538235",

volume = "2",

pages = "358--365",

booktitle = "Proceedings - 12th IEEE International Conference on Computational Science and Engineering, CSE 2009 - 7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009",

note = "7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009 ; Conference date: 29-08-2009 Through 31-08-2009",

}

Yang, Y, Wang, M, Shao, Z & Guo, M 2009, Dynamic scratch-pad memory management with data pipelining for embedded systems. in Proceedings - 12th IEEE International Conference on Computational Science and Engineering, CSE 2009 - 7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009. vol. 2, 5284058, pp. 358-365, 7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009, Vancouver, BC, Canada, 29/08/09. https://doi.org/10.1109/CSE.2009.295

Dynamic scratch-pad memory management with data pipelining for embedded systems. / Yang, Yanqin; Wang, Meng; Shao, Zili et al.
Proceedings - 12th IEEE International Conference on Computational Science and Engineering, CSE 2009 - 7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009. Vol. 2 2009. p. 358-365 5284058.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Dynamic scratch-pad memory management with data pipelining for embedded systems

AU - Yang, Yanqin

AU - Wang, Meng

AU - Shao, Zili

AU - Guo, Minyi

PY - 2009/12/3

Y1 - 2009/12/3

N2 - In this paper, we propose an effective data pipelining technique, SPDP (scratch-pad data pipelining), for dynamic scratch-pad memory (SPM) management with DMA (Direct Memory Access). In SPDP, we group multiple iterations of a loop into a block for SPM allocation, and implement a data pipeline by overlapping the execution of CPU instructions and DMA operations. We have implemented our SPDP technique into the IMPACT compiler, and conduct experiments using a set of benchmarks from DSPstone, Mibench and Mediabench on the cycle-accurate VLIW simulator of Trimaran. The experimental results show that our technique achieves significant performance improvement compared with the previous work.

AB - In this paper, we propose an effective data pipelining technique, SPDP (scratch-pad data pipelining), for dynamic scratch-pad memory (SPM) management with DMA (Direct Memory Access). In SPDP, we group multiple iterations of a loop into a block for SPM allocation, and implement a data pipeline by overlapping the execution of CPU instructions and DMA operations. We have implemented our SPDP technique into the IMPACT compiler, and conduct experiments using a set of benchmarks from DSPstone, Mibench and Mediabench on the cycle-accurate VLIW simulator of Trimaran. The experimental results show that our technique achieves significant performance improvement compared with the previous work.

UR - https://www.scopus.com/pages/publications/70749112938

U2 - 10.1109/CSE.2009.295

DO - 10.1109/CSE.2009.295

M3 - Conference article published in proceeding or book

SN - 9780769538235

VL - 2

SP - 358

EP - 365

BT - Proceedings - 12th IEEE International Conference on Computational Science and Engineering, CSE 2009 - 7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009

T2 - 7th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2009

Y2 - 29 August 2009 through 31 August 2009

ER -

Dynamic scratch-pad memory management with data pipelining for embedded systems

Abstract

Conference

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this