One of the Internet’s fastest-growing technologies is cloud computing, which needs each user to share various resources and virtual computers to access the services they demand. The difficulty is in allocating the user’s tasks among the many resources at hand while maintaining a balanced workload. In this work, we present the design of a new scheduling approach for cloud virtual machines in order to process events (related to resource overuse or underuse). This reduces the amount of effort consumed while maintaining the required performance levels in the cloud data center. To evaluate the effectiveness of the approach with real load paths, we conducted an experiment using the open-source cloud computing platform OpenNebula.