Application of a Task Stalling Buffer in Distributed Hybrid Cloud Computing

Abstract

The purpose of the research is to create a hybrid cloud platform that performs distributed computing tasks using high-performance servers and volunteer computing resources. The proposed platform uses a new task scheduling method, which is also presented in this paper. It uses a task stalling buffer to manage workload among the two grids without any additional information about the tasks. Since efficient task scheduling in these distributed systems is the actual problem, the system reliability issue is solved using a hybrid cloud architecture when both high-performance servers and volunteer computing resources are combined. The results of the experiment showed that the proposed solution solves the problem of balancing workload between two grids better than the standard scheduling algorithm. Computer study and experiments also showed that the proposed hybrid cloud tasks scheduling method with a task stalling buffer reduces up to 47.3 % of total task execution time. The outcome of this paper provides a background for future research on a task stalling buffer in hybrid cloud computing.