Global variable partition with virtually shared scratch pad memory to minimize schedule length

Meikang Qiu; Lei Zhang; Minyi Guo; Fei Hu; Shaobo Liu; Edwin H.M. Sha

doi:10.1109/ICPPW.2009.22

Global variable partition with virtually shared scratch pad memory to minimize schedule length

Meikang Qiu^*, Lei Zhang, Minyi Guo, Fei Hu, Shaobo Liu, Edwin H.M. Sha

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

On-chip memory is a critical component of MPSoC architecture. Virtually Shared Scratch Pad Memory (VSSPM) architecture is a novel approach to close the processormemory speed gap since it can take the advantages of both shared and private SPM. In this paper, we propose an efficient variable partitioning algorithms, Global View Prediction (GVP), to improve the overall throughput. The experimental results demonstrate the effectiveness of our approach.

Original language	English
Title of host publication	ICPPW 2009 - The 38th International Conference Parallel Processing Workshops
Pages	478-483
Number of pages	6
DOIs	https://doi.org/10.1109/ICPPW.2009.22
State	Published - 2009
Externally published	Yes
Event	38th International Conference Parallel Processing Workshops, ICPPW 2009 - Vienna, Austria Duration: 22 Sep 2009 → 25 Sep 2009

Publication series

Name	Proceedings of the International Conference on Parallel Processing Workshops
ISSN (Print)	1530-2016

Conference

Conference	38th International Conference Parallel Processing Workshops, ICPPW 2009
Country/Territory	Austria
City	Vienna
Period	22/09/09 → 25/09/09

Access to Document

10.1109/ICPPW.2009.22

Cite this

Qiu, M., Zhang, L., Guo, M., Hu, F., Liu, S., & Sha, E. H. M. (2009). Global variable partition with virtually shared scratch pad memory to minimize schedule length. In ICPPW 2009 - The 38th International Conference Parallel Processing Workshops (pp. 478-483). Article 5365342 (Proceedings of the International Conference on Parallel Processing Workshops). https://doi.org/10.1109/ICPPW.2009.22

@inproceedings{6fd1ee491197462c9e6c7c87945af9eb,

title = "Global variable partition with virtually shared scratch pad memory to minimize schedule length",

abstract = "On-chip memory is a critical component of MPSoC architecture. Virtually Shared Scratch Pad Memory (VSSPM) architecture is a novel approach to close the processormemory speed gap since it can take the advantages of both shared and private SPM. In this paper, we propose an efficient variable partitioning algorithms, Global View Prediction (GVP), to improve the overall throughput. The experimental results demonstrate the effectiveness of our approach.",

author = "Meikang Qiu and Lei Zhang and Minyi Guo and Fei Hu and Shaobo Liu and Sha, \{Edwin H.M.\}",

year = "2009",

doi = "10.1109/ICPPW.2009.22",

language = "英语",

isbn = "9780769538037",

series = "Proceedings of the International Conference on Parallel Processing Workshops",

pages = "478--483",

booktitle = "ICPPW 2009 - The 38th International Conference Parallel Processing Workshops",

note = "38th International Conference Parallel Processing Workshops, ICPPW 2009 ; Conference date: 22-09-2009 Through 25-09-2009",

}

Qiu, M, Zhang, L, Guo, M, Hu, F, Liu, S & Sha, EHM 2009, Global variable partition with virtually shared scratch pad memory to minimize schedule length. in ICPPW 2009 - The 38th International Conference Parallel Processing Workshops., 5365342, Proceedings of the International Conference on Parallel Processing Workshops, pp. 478-483, 38th International Conference Parallel Processing Workshops, ICPPW 2009, Vienna, Austria, 22/09/09. https://doi.org/10.1109/ICPPW.2009.22

Global variable partition with virtually shared scratch pad memory to minimize schedule length. / Qiu, Meikang; Zhang, Lei; Guo, Minyi et al.
ICPPW 2009 - The 38th International Conference Parallel Processing Workshops. 2009. p. 478-483 5365342 (Proceedings of the International Conference on Parallel Processing Workshops).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Global variable partition with virtually shared scratch pad memory to minimize schedule length

AU - Qiu, Meikang

AU - Zhang, Lei

AU - Guo, Minyi

AU - Hu, Fei

AU - Liu, Shaobo

AU - Sha, Edwin H.M.

PY - 2009

Y1 - 2009

N2 - On-chip memory is a critical component of MPSoC architecture. Virtually Shared Scratch Pad Memory (VSSPM) architecture is a novel approach to close the processormemory speed gap since it can take the advantages of both shared and private SPM. In this paper, we propose an efficient variable partitioning algorithms, Global View Prediction (GVP), to improve the overall throughput. The experimental results demonstrate the effectiveness of our approach.

AB - On-chip memory is a critical component of MPSoC architecture. Virtually Shared Scratch Pad Memory (VSSPM) architecture is a novel approach to close the processormemory speed gap since it can take the advantages of both shared and private SPM. In this paper, we propose an efficient variable partitioning algorithms, Global View Prediction (GVP), to improve the overall throughput. The experimental results demonstrate the effectiveness of our approach.

UR - https://www.scopus.com/pages/publications/77949503383

U2 - 10.1109/ICPPW.2009.22

DO - 10.1109/ICPPW.2009.22

M3 - 会议稿件

AN - SCOPUS:77949503383

SN - 9780769538037

T3 - Proceedings of the International Conference on Parallel Processing Workshops

SP - 478

EP - 483

BT - ICPPW 2009 - The 38th International Conference Parallel Processing Workshops

T2 - 38th International Conference Parallel Processing Workshops, ICPPW 2009

Y2 - 22 September 2009 through 25 September 2009

ER -

Global variable partition with virtually shared scratch pad memory to minimize schedule length

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this