CAJ | 학술논문

针对已有云计算任务调度算法为实现最短时间跨度而不能兼顾负载均衡和服务质量的问题，提出基于遗传算法和蚁群算法融合的QoS约束任务调度策略CAAC。CAAC利用任务的预测完成时间和成本耗费定义适应度函数；通过遗传算子全局搜索最优解，融合蚁群算子提高解的精确度；当任务数量大于50时，该算法收敛速度和资源利用率比蚁群算法平均提高4.7'和30.8'。仿真结果表明，该算法在保证服务质量和资源负载均衡方面具有优越性。
침대이유운계산임무조도산법위실현최단시간과도이불능겸고부재균형화복무질량적문제，제출기우유전산법화의군산법융합적QoS약속임무조도책략CAAC。CAAC이용임무적예측완성시간화성본모비정의괄응도함수；통과유전산자전국수색최우해，융합의군산자제고해적정학도；당임무수량대우50시，해산법수렴속도화자원이용솔비의군산법평균제고4.7'화30.8'。방진결과표명，해산법재보증복무질량화자원부재균형방면구유우월성。
In current environment of cloud computing, some task scheduling algorithms are in the purpose of achieving the shortest time span and ingoring the load balance and quality of service. To solve the problem, the thesis proposed CAAC, a task scheduling strategy with Quality of Service ( QoS) constraints based on genetic algorithm and ant colony algorithm. The new algorithm made use of the predict cost of time and money to definite fitness function, searching the optimal solution in the global domain through the genetic operator and combining ant operator to improve the accuracy of solution. When the scale of tasks is greater than 50, compared with ant colony algorithm, CAAC' s convergence and utilization can be improved by 4. 7' and 30. 8' respectively. The simulation results show that the algorithm is more efficient in balancing resources load and guaranting QoS.