Dynamic Level Task Scheduling Algorithm based on Risk Estimation Model in Grid Computing

In a heterogeneous distributed Grid system, machines and network failures are inevitable and can have an adverse effect on applications execution on the system, therefore the objective of task scheduling is not only to minimize the makespan of applications, but also to reduce risks for application execution failures. A new algorithm named by Risk-DLS algorithm will be proposed, which will combine a risk estimation model of resources with the DLS algorithm through an improved equation of dynamic level. The simulation shows that when choosing reasonable parameters, Risk-DLS algorithm is better than DLS algorithm in makespan, at the same time, it can also reduce the adverse effect due to uncertain environment of grid computing.