PROJECT TITLE :

Scheduling Algorithms for Efficient Execution of Stream Workflow Applications in Multicloud Environments

ABSTRACT:

The applications used for processing large amounts of data are becoming increasingly complicated. They are no longer of a monolithic nature; rather, in their place is a workflow that is composed of analytical processes that are decoupled from one another. Among these workflow applications, the stream workflow application is one type that integrates numerous streaming big data applications to provide decision-making support. Each analytic subcomponent of these applications operates in a non-stop loop and processes data streams at a rate that is determined by a number of factors, including the amount of bandwidth available on the network and the processing speed of the parent analytical subcomponent. Because of this, the execution of these applications in cloud environments calls for the utilization of sophisticated scheduling methods that are compliant with the requirements of the end user in terms of the data processing and the time limit for making decisions. In this article, we propose two multicloud scheduling and resource allocation techniques for efficiently executing stream workflow applications on multicloud environments while adhering to workflow application and user performance requirements and reducing the cost of execution. These techniques aim to minimize the amount of money spent on the execution process. The findings demonstrated that the genetic algorithm that was proposed is both suitable and efficient for all of the experiments.