sis 19(22): e7

Research Article

Improving Performance in Component Based Distributed Systems

Download853 downloads
  • @ARTICLE{10.4108/eai.13-7-2018.159357,
        author={Fahd N.  Al-Wesabi and Huda G.  Iskandar and Mokhtar M. Ghilan},
        title={Improving Performance in Component Based Distributed Systems},
        journal={EAI Endorsed Transactions on Scalable Information Systems},
        volume={6},
        number={22},
        publisher={EAI},
        journal_a={SIS},
        year={2019},
        month={7},
        keywords={Distributed systems, component based distributed systems, performance, availability and delay time},
        doi={10.4108/eai.13-7-2018.159357}
    }
    
  • Fahd N. Al-Wesabi
    Huda G. Iskandar
    Mokhtar M. Ghilan
    Year: 2019
    Improving Performance in Component Based Distributed Systems
    SIS
    EAI
    DOI: 10.4108/eai.13-7-2018.159357
Fahd N. Al-Wesabi1,*, Huda G. Iskandar2, Mokhtar M. Ghilan3
  • 1: Assistant professor, College of Science and Arts in Mahayel Asir, King Khalid University, KSA, Faculty of Computer and IT, Sana’a University, Sana’a, Yemen
  • 2: Master candidate, Faculty of Computer and IT, Sana’a University, Sana’a, Yemen
  • 3: Assistant professor, Faculty of Computer and IT, Sana’a University, Sana’a, Yemen
*Contact email: Fwesabi@gmail.com

Abstract

Assuring high performance is one of the most important factors when building computerized systems. In distributed component-based systems different configurations and workloads are common. Over the years, many approaches were proposed to analyse, predict, measure and evaluate the performance of component-based distributed systems in an attempt to improve its performance. Higher performance results in better utilization of system resources, high throughput and quick response time of user requests. In this paper, we extend the work of the E-Avala approach to improve the overall performance of the proposed approach by adding power processing capabilities. The proposed approach has been implemented and compared with previous approach using Arena simulation with varying configuration parameters. Comparative results show that the proposed approach has given better performance with different levels of processing powers.