skip to main content
10.1145/3641584.3641795acmotherconferencesArticle/Chapter ViewAbstractPublication PagesaiprConference Proceedingsconference-collections
research-article

A PSOTP Algorithm Based on Static Task Allocation of Neural Network

Published: 14 June 2024 Publication History

Abstract

With the rise of heterogeneous computing, the problem of task allocation on heterogeneous platforms consisting of multiple architectures has become a major research hotspot in heterogeneous computing. The particle swarm optimization (PSO) algorithm has good performance in solving optimal problems with strong universality and fast convergence speed. However, different initial particles can lead to varying search times, and particles are prone to falling into local optima during the search process. Greedy algorithms can approach the goal faster by starting from the initial solution of the problem when solving it. Therefore, this paper combines the greedy algorithm to set the initial particles of the PSO algorithm, which shortens the particle search time. The inclusion of random perturbation terms avoids falling into local optima. A particle swarm optimization task partitioning (PSOTP) algorithms is designed, which can statically allocate tasks to neural networks in a CPU-GPU heterogeneous computing architectures. The experimental results show that PSOTP algorithms can reduce the time to search for the best scheme by 22% compared with the PSO algorithm, and the accuracy of the search scheme is higher.

References

[1]
Bouvier D, Cohen B, Fry W, Kabini: An AMD Accelerated Processing Unit System on A Chip[J]. IEEE Micro, 2014, 34(2):22-33.
[2]
Jouppi N P, Young C, Patil N, In-Datacenter Performance Analysis of a Tensor Processing Unit[C]// the 44th Annual International Symposium. IEEE Computer Society, 2017:1-12.
[3]
Ouyang J, Du X, Ma Y, 3.3 Kunlun: A 14nm High-Performance AI Processor for Diversified Workloads[C]// 2021 IEEE International Solid- State Circuits Conference (ISSCC). IEEE, 2021.
[4]
[4] Liu H, Ma H, Liu Q, An efficient and robust GPGPU-parallelized contact algorithm for the combined finite-discrete element method[J]. Computer Methods in Applied Mechanics and Engineering, 2022(May 15):395.
[5]
B. N. Chandrashekar, K. Aditya Shastry, B. A. Manjunath and V. Geetha, "Performance Model of HPC Application On CPU-GPU Platform," 2022 IEEE 2nd Mysore Sub Section International Conference (MysuruCon), Mysuru, India, 2022, pp. 1-6. W, Zhou X, Zhou J T Y, Accelerating attention mechanism on fpgas based on efficient reconfigurable systolic array[J]. ACM Transactions on Embedded Computing Systems (TECS), 2022.
[6]
C. Gong, M. Ma, L. Wu, W. Liu, Y. Zhou and Y. Yang, "Task Offloading and Resource Allocation in CPU-GPU Heterogeneous Networks," GLOBECOM 2022 - 2022 IEEE Global Communications Conference, Rio de Janeiro, Brazil, 2022, pp. 4492-4497.
[7]
R. Kang, H. Li, W. Li and Y. Zhou, "A Novel PSO Approach for Cooperative Task Assignment of Multi-UAV Attacking Moving Targets," 2022 34th Chinese Control and Decision Conference (CCDC), Hefei, China, 2022, pp. 3670-3675.
[8]
T. Issac, S. Silas and E. B. Rajsingh, "Investigations on PSO based task assignment algorithms for heterogeneous wireless sensor network," 2019 2nd International Conference on Signal Processing and Communication (ICSPC), Coimbatore, India, 2019, pp. 89-93.
[9]
Y. -z. Zhang, J. -w. Li, B. Hu and J. -d. Zhang, "An improved PSO algorithm for solving multi-UAV cooperative reconnaissance task decision-making problem," 2016 IEEE International Conference on Aircraft Utility Systems (AUS), Beijing, China, 2016, pp. 434-437.
[10]
Kennedy J, Eberhart R . Particle Swarm Optimization[C]// Icnn95-international Conference on Neural Networks. IEEE, 1995.
[11]
S. Rawat and M. K. Khandewal, "Improved Multiphase PSO using Greedy Approach for Effective Population Size," 2023 5th International Conference on Smart Systems and Inventive Technology (ICSSIT), Tirunelveli, India, 2023, pp. 1626-1632.
[12]
H. Zhai, W. Wang, Q. Li and W. Zhang, "Weapon-Target Assignment Based on Improved PSO Algorithm," 2021 33rd Chinese Control and Decision Conference (CCDC), Kunming, China, 2021, pp. 6320-6325.
[13]
D. Dev Misra, K. K. Sarma, U. Bhattacharjee, P. K. Goswami and N. Mastorakis, "Optimal Routing in the 5G Ultra Dense Small Cell Network using GA, PSO and Hybrid PSO-GA Evolutionary Algorithms," 2020 24th International Conference on Circuits, Systems, Communications and Computers (CSCC), Chania, Greece, 2020, pp. 39-44.
[14]
Shi Y . A Modified Particle Swarm Optimizer[C]// Proc of IEEE Icec Conference. 1998.(CCDC),Kunming,China,2021,pp.6320-6325.
[15]
Zhang X, Du Y, Zheng Q, A Modified Particle Swarm Optimizer for Tracking Dynamic Systems[J]. Springer Berlin Heidelberg, 2005.
[16]
K. S. Gajjala and B. Chakraborty, "Human Activity Recognition based on LSTM Neural Network Optimized by PSO Algorithm," 2021 IEEE 4th International Conference on Knowledge Innovation and Invention (ICKII), Taichung, Taiwan, 2021, pp. 128-133. Wiedemann,

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Other conferences
AIPR '23: Proceedings of the 2023 6th International Conference on Artificial Intelligence and Pattern Recognition
September 2023
1540 pages
ISBN:9798400707674
DOI:10.1145/3641584
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 14 June 2024

Permissions

Request permissions for this article.

Check for updates

Author Tags

  1. Heterogeneous computing
  2. Neural Network
  3. PSO
  4. Task Allocation

Qualifiers

  • Research-article
  • Research
  • Refereed limited

Conference

AIPR 2023

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • 0
    Total Citations
  • 17
    Total Downloads
  • Downloads (Last 12 months)17
  • Downloads (Last 6 weeks)2
Reflects downloads up to 10 Feb 2025

Other Metrics

Citations

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

HTML Format

View this article in HTML Format.

HTML Format

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media