QoS-constrained policy quorum configuration scheme for reliable resource management in grid

Abstract

Due to the large scale of Grid environment and its rapid expansion, both in Grid resources and in Grid network environment, it is getting more imperative to provide a Grid management mechanism that enables Grid Computing to adapt to various application requirements in a flexible and automated way. As efficient resource management, Policy-based Management (PBM) is very suitable to such a complex management environment; however the current PBM architecture has problems in the sense that it can only address fairly limited issues and usually requires human intervention. In this thesis, we propose QoS-constrained Policy Quorum configuration scheme for Reliable Resource Management. QoS-constrained Policy has an advantage to guarantee minimal QoS requirements requested by users. A session admission controller (SAC) in policy based management architecture is responsible for providing QoS service to users by checking for availability of resources in compliance to SLAs, before starting the requested application session. Generally, Quorum systems are well-known tools for ensuring the consistency and availability of replicated data despite the benign failure of data repositories. The intersection between Quorums has the innate fault-tolerable property. Hence the Quorum based management architecture can support a reliable region of various resources. In order to apply the Quorum based model to our scheme, we define elements in Quorum as QoS based middleware entities and network entities, represented as a binary form. In our Quorum based model, QoS Quorum means the condition including QoS entities required by user and Resource Quorum means the resource including entities that can satisfy the condition of QoS Quorum. Therefore Resource Quorum, mapped with QoS Quorum, can guarantee user’s QoS requirements. We also suggest a heuristic configuration algorithm in order to optimize performance and usage cost of Resource Quorum. We carry out both simulations and experiments...