Real-time performance analysis of distributed multithreaded applications in a cluster of ARM-based embedded devices
by George K. Adam
International Journal of High Performance Systems Architecture (IJHPSA), Vol. 11, No. 2, 2022

Abstract: The challenges in real-time cluster computing, particularly in computing efficiency and reliability, have evolved significantly due to the increase of Internet of Things (IoT), cloud and edge computing applications. Lately, a number of low-power and low-cost clusters have appeared, based upon single board computers, which deploy multithreading techniques to run in parallel thousands of tasks, to support the ever-increasing demand for timely data processing. However, their real-time performance is still under research. This paper proposes a real-time performance analysis approach for evaluating metrics such as thread execution time, response time, parallel efficiency and speedup. The measurements are based upon recursively generated multithreaded applications running in parallel and distributed across multiple cores within a cluster of Raspberry Pi4, running Linux with real-time support and interconnected on a fast gigabit Ethernet. The experimental results validate the efficiency of the proposed approach and show that real-time support enables higher throughput across all workloads, and lower execution times.

Online publication date: Fri, 16-Dec-2022

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of High Performance Systems Architecture (IJHPSA):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com