Automated Dynamic Service Placement And Replication Framework Using Team Formation Approach To Enhance Service Availability
Loading...
Date
2012-05
Authors
Ooi, Boon Yaik
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
Managing and administering services in the distributed environment are getting more complicated as the size of the distributed computing environment grows larger and becomes more dynamic. This thesis proposes an automated framework to manage services in a dynamic distributed environment where the resources available for these services would change from time to time. The aim of this research is to design an automated framework that would continuously search for resources with better performance and pool resources together to achieve better availability using a team formation approach. Human administrators are still required but are freed from making low-level decisions such as to decide the actual placement of the services and to design the failover procedures for each of the services. In addition to that, this research also introduces a resource evaluation method that fused Fuzzy Logic (FL) and Adaptive Network-based Fuzzy Inference System (ANFIS). It uses FL to capture services’ requirements from administrators and ANFIS to allow administrators to fine-tune the resources evaluation process when environment resources change. Simulations were developed to evaluate the ability of the proposed framework in improving service availability using multiple consecutive random fault injection method. The experimental results showed that the framework can improve service availability even during multiple consecutive resource failure. Besides that, this research also highlights the limitations and potential future enhancement of the proposed work.
Description
Keywords
Replication framework using team formation , approach to enhance service availability