A Hybrid QoS-Based Algorithm for Independent Tasks Scheduling in Grid

According to QoS (quality of service) of computing resources in grid, a 2-tier hybrid algorithm is proposed for independent tasks scheduling with time-cost minimization. QGMM (QoS Guided Min-Min, presented by He Xiao Shan et al., Journal of Computer Science and Technology, 2003) is adopted in the first level to obtain a schedule with optimized completion time. A GA (genetic algorithm) only with mutation operator is used to optimize cost with the same obtained completion time in the second tier. The proposed approach is compared with Min-Min and QGMM, efficient existing heuristics for independent tasks scheduling with completion time minimization. Experimental results show that the proposed approach can get lower cost than the other two with the same completion time as QGMM, which is better than that of Min-Min