A novel SLA-driven verification and planning framework for QoS

QoS (Quality of Service) guarantee has emerged as one of the most important technologies for the grid computing. Because the existing QoS control systems focus on designing independent executive management system, some problems still remain to be addressed, including low efficiency, bad scalability, and inadequate QoS automatic control. To address the above problems, this paper presents a novel SLA-driven system framework with verification and planning for QoS. The framework integrates functions of executive management with system services, achieves the optimization of system monitoring & management and automatic QoS control through a novel automatic combination of services, and scales well. The system framework with verification and planning for QoS is applied to heuristic scheduling to improve the Min-Min algorithm. An illustrative example and experimental results show that the system framework with verification and planning for QoS can effectively improve the efficiency of monitoring and management for QoS and facilitate automatic QoS control.