Accelerated P systems with active membranes on a graphics processing unit

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

Abstract

P systems have parallel structures; therefore, parallel tools like graphic processing units (GPU) are needed to extract this parallelism. In previous approaches each membrane was assigned to one thread block. This was inefficient when the number of objects inside each membrane was low. In the proposed approach objects of the membranes are represented as entries of a matrix. Then this matrix can be divided in sub blocks (that can include more than one membrane) to run on thread blocks of GPU efficiently. With this approach, the number of active threads in each thread block can be balanced and processing speed enhanced. For example, using previous approaches, for two objects in each membrane since one membrane is assigned to one thread block the multiprocessor occupancy is low and the speedup is 0.6 times, whereas the proposed approach multiprocessor occupancy is high and achieves a 33.7-times speedup with respect to sequential implementation.

Original languageEnglish
Title of host publicationProceedings - Asian Conference on Membrane Computing, ACMC 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Print)9781479980123
DOIs
Publication statusPublished - 23 Mar 2014
Event2014 Asian Conference on Membrane Computing, ACMC 2014 - Coimbatore, Tamil Nadu, India
Duration: 18 Sep 201419 Sep 2014

Other

Other2014 Asian Conference on Membrane Computing, ACMC 2014
CountryIndia
CityCoimbatore, Tamil Nadu
Period18/9/1419/9/14

Fingerprint

Membranes
Graphics processing unit
Processing

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Science Applications

Cite this

Maroosi, A., & Muniyandi, R. C. (2014). Accelerated P systems with active membranes on a graphics processing unit. In Proceedings - Asian Conference on Membrane Computing, ACMC 2014 [7065803] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ACMC.2014.7065803

Accelerated P systems with active membranes on a graphics processing unit. / Maroosi, Ali; Muniyandi, Ravie Chandren.

Proceedings - Asian Conference on Membrane Computing, ACMC 2014. Institute of Electrical and Electronics Engineers Inc., 2014. 7065803.

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

Maroosi, A & Muniyandi, RC 2014, Accelerated P systems with active membranes on a graphics processing unit. in Proceedings - Asian Conference on Membrane Computing, ACMC 2014., 7065803, Institute of Electrical and Electronics Engineers Inc., 2014 Asian Conference on Membrane Computing, ACMC 2014, Coimbatore, Tamil Nadu, India, 18/9/14. https://doi.org/10.1109/ACMC.2014.7065803
Maroosi A, Muniyandi RC. Accelerated P systems with active membranes on a graphics processing unit. In Proceedings - Asian Conference on Membrane Computing, ACMC 2014. Institute of Electrical and Electronics Engineers Inc. 2014. 7065803 https://doi.org/10.1109/ACMC.2014.7065803
Maroosi, Ali ; Muniyandi, Ravie Chandren. / Accelerated P systems with active membranes on a graphics processing unit. Proceedings - Asian Conference on Membrane Computing, ACMC 2014. Institute of Electrical and Electronics Engineers Inc., 2014.
@inproceedings{835e02e4a84845c180f8ef6230e9509a,
title = "Accelerated P systems with active membranes on a graphics processing unit",
abstract = "P systems have parallel structures; therefore, parallel tools like graphic processing units (GPU) are needed to extract this parallelism. In previous approaches each membrane was assigned to one thread block. This was inefficient when the number of objects inside each membrane was low. In the proposed approach objects of the membranes are represented as entries of a matrix. Then this matrix can be divided in sub blocks (that can include more than one membrane) to run on thread blocks of GPU efficiently. With this approach, the number of active threads in each thread block can be balanced and processing speed enhanced. For example, using previous approaches, for two objects in each membrane since one membrane is assigned to one thread block the multiprocessor occupancy is low and the speedup is 0.6 times, whereas the proposed approach multiprocessor occupancy is high and achieves a 33.7-times speedup with respect to sequential implementation.",
author = "Ali Maroosi and Muniyandi, {Ravie Chandren}",
year = "2014",
month = "3",
day = "23",
doi = "10.1109/ACMC.2014.7065803",
language = "English",
isbn = "9781479980123",
booktitle = "Proceedings - Asian Conference on Membrane Computing, ACMC 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Accelerated P systems with active membranes on a graphics processing unit

AU - Maroosi, Ali

AU - Muniyandi, Ravie Chandren

PY - 2014/3/23

Y1 - 2014/3/23

N2 - P systems have parallel structures; therefore, parallel tools like graphic processing units (GPU) are needed to extract this parallelism. In previous approaches each membrane was assigned to one thread block. This was inefficient when the number of objects inside each membrane was low. In the proposed approach objects of the membranes are represented as entries of a matrix. Then this matrix can be divided in sub blocks (that can include more than one membrane) to run on thread blocks of GPU efficiently. With this approach, the number of active threads in each thread block can be balanced and processing speed enhanced. For example, using previous approaches, for two objects in each membrane since one membrane is assigned to one thread block the multiprocessor occupancy is low and the speedup is 0.6 times, whereas the proposed approach multiprocessor occupancy is high and achieves a 33.7-times speedup with respect to sequential implementation.

AB - P systems have parallel structures; therefore, parallel tools like graphic processing units (GPU) are needed to extract this parallelism. In previous approaches each membrane was assigned to one thread block. This was inefficient when the number of objects inside each membrane was low. In the proposed approach objects of the membranes are represented as entries of a matrix. Then this matrix can be divided in sub blocks (that can include more than one membrane) to run on thread blocks of GPU efficiently. With this approach, the number of active threads in each thread block can be balanced and processing speed enhanced. For example, using previous approaches, for two objects in each membrane since one membrane is assigned to one thread block the multiprocessor occupancy is low and the speedup is 0.6 times, whereas the proposed approach multiprocessor occupancy is high and achieves a 33.7-times speedup with respect to sequential implementation.

UR - http://www.scopus.com/inward/record.url?scp=84946690562&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84946690562&partnerID=8YFLogxK

U2 - 10.1109/ACMC.2014.7065803

DO - 10.1109/ACMC.2014.7065803

M3 - Conference contribution

AN - SCOPUS:84946690562

SN - 9781479980123

BT - Proceedings - Asian Conference on Membrane Computing, ACMC 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -