An improved task scheduling and load balancing algorithm under the heterogeneous cloud computing network

In recent decades, with the rapid development and popularization of Internet and computer technology, cloud computing had become a highly-demanded service due to the advantages of high computing power, cheap cost of services, scalability, accessibility as well as availability. However, a fly in the ointment was that the system is more complex while dispatching variety of tasks to servers. It means that dispatching tasks to the servers is a challenge since there has a large number of heterogeneous servers, core and diverse application services need to cooperate with each other in the cloud computing network. To deal with the huge number of tasks, an appropriate and effective scheduling algorithm is to allocate these tasks to appropriate servers within the minimum completion time, and to achieve the load balancing of workload. Based on the reasons above, a novel dispatching algorithm, called Advanced MaxSufferage algorithm (AMS), is proposed in this paper to improve the dispatching efficiency in the cloud computing network. The main concept of the AMS is to allocate the tasks to server nodes by comparing the SV value, MSV value, and average value of expected completion time of the server nodes between each task. Basically, the AMS algorithm can obtain better task completion time than previous works and can achieve loadbalancing in cloud computing network.