Semantic-based QoS management in cloud systems: Current status and future challenges
Future Generation Computer Systems
| Hi-index | 0.00 |
The need of mechanisms to guarantee the Quality of Service (QoS) in Grid environments has empowered the interest on Service Level Agreements (SLAs), which represent the contract established by negotiation between users and service providers that must be accomplished. Consequently, the SLAs contain numerous terms related to the QoS expected. The aim of this paper is to address different QoS levels in Grid environments by using specific metrics on SLAs. These metrics represent the expected QoS by the user for the given service level terms. Hence, a characterization and classification of these terms, making a differentiation of QoS in levels, is needed in order to improve the number of fulfilled agreements. Our proposal is based on a WS-Agreement compliant architecture developed to provide support for SLAs in Grid environments, also providing a framework prepared for new terms and policy definitions. More precisely, the QoS expected by users is clearly defined in three levels. These levels are used to classify the compromise adquired by each SLA (depending on the expectations of the user that submitted it) and deal with the confidence that Grid resources provide. The evaluation on a real Grid testbed shows the efficiency of the proposal.