Byzantine Attack Identification via Group Testing Approach in Coded Distributed Computing for Matrix Multiplication 


Vol. 48,  No. 2, pp. 249-257, Feb.  2023
10.7840/kics.2023.48.2.249


PDF
  Abstract

In a distributed computing system utilizing multiple computing nodes, coded distributed computing has been proposed to reduce straggler effects, ensure data privacy, and so on. However, due to the characteristic of distributed computing systems where many computing nodes called workers are utilized, there may exist Byzantine workers that carry out malicious attacks by sending wrong computation results to contaminate the overall computation results. In this paper, for the case where some Byzantine workers that return wrong computation results exist in coded distributed computing for matrix multiplication tasks, we introduce a new coded distributed computing scheme for effective identifying Byzantine workers and decoding the overall computation results. First, we suggest a new coded distributed computing technique for identifying Byzantine workers, and propose two group testing approaches suitable for the proposed coded distributed computing scheme. In addition, we show the Byzantine worker identification time to identify Byzantine workers via experiments that perform matrix multiplication tasks in the presence of Byzantine workers in Amazon Web Service (AWS) distributed computing systems.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

S. Hong, H. Yang, Y. Yoon, J. Lee, "Byzantine Attack Identification via Group Testing Approach in Coded Distributed Computing for Matrix Multiplication," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 2, pp. 249-257, 2023. DOI: 10.7840/kics.2023.48.2.249.

[ACM Style]

Sangwoo Hong, Heecheol Yang, Youngseok Yoon, and Jungwoo Lee. 2023. Byzantine Attack Identification via Group Testing Approach in Coded Distributed Computing for Matrix Multiplication. The Journal of Korean Institute of Communications and Information Sciences, 48, 2, (2023), 249-257. DOI: 10.7840/kics.2023.48.2.249.

[KICS Style]

Sangwoo Hong, Heecheol Yang, Youngseok Yoon, Jungwoo Lee, "Byzantine Attack Identification via Group Testing Approach in Coded Distributed Computing for Matrix Multiplication," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 2, pp. 249-257, 2. 2023. (https://doi.org/10.7840/kics.2023.48.2.249)
Vol. 48, No. 2 Index

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880