Cloud Computing is quickly developing and a lot more cloud suppliers are arising. This project says how the incoming tasks are scheduled in the cloud. In this paper, presenting a classifier algorithmic technique that integrates the K-Nearest Neighbor(KNN) algorithm with the Naive Bayes algorithm. Calculation offloading innovation offers a feasible arrangement by offloading some calculation serious assignments of the K-Nearest Neighbor algorithm and the Naive Bayes algorithm to edges or distant mists that are furnished with adequate assets. However, the offloading process might lead to excessive delays and thus seriously affect the user experience. To address this significant issue, we first respect the average response time of multi-task parallel scheduling as our streamlining objective. Finally, the K-Nearest Neighbor and Naive Bayes algorithm based applications are proposed to solve the problem. Offloading a task means how the task is scheduled in Virtual Machine, first, the task is separated into types(small, medium, large and extra-large) after that it sees which Virtual Machine is available to load the task. Suppose, if the incoming task is too-large then we have to offload it by separating the task into subtask as 6 operations (split, combine, merge, demerge, promote and demote). To avoid the burden in the Virtual Machine, We analyze the above operations by using the K-Nearest Neighbour (KNN) algorithm with the Naive Bayes algorithm and then scheduled in the Virtual Machine. We want to show that the K-Nearest Neighbour (KNN) algorithm with the Naive Bayes algorithm shows better performance than Deep Neural Network(DNN).