Workflow scheduling optimisation for distributed environment using artificial neural networks and reinforcement learning
by K. Jairam Naik; Mounish Pedagandam; Amrita Mishra
International Journal of Computational Science and Engineering (IJCSE), Vol. 24, No. 6, 2021

Abstract: The growing volumes of information and multifaceted nature of information processing, workflow scheduling in distributed environment are a prominent component for computing operations to diminish the amount of information transferring, computation load allocation to resources, reducing the task's waiting time and execution time. The basic objective of this article is to find an optimal schedule (Sopt) which can reduce the makespan of workflow. Artificial intelligence and neural network (NN) systems are the main-stream, but they were not effectively employed nevertheless for workflow scheduling. Hence, we enhance the scheduling by realising artificial neural networks and reinforcement Q-learning standards. An optimised NN-based scheduling algorithm (WfSo_ANRL) that represents an agent which can effectively schedule the tasks among computational nodes was provided in this article. The agent interacts with the external environment, i.e., the computing environment and collects the current status of load encoded in the form of a state vector. The agent then predicts an action and efficiently allocates the tasks on to the attainable resources. The external computing environment then awards incentives to the agent. The agent then learns to produce optimal schedules for reducing the makespan. In this way, the WfSo_ANRL produces optimal solution for workload.

Online publication date: Tue, 04-Jan-2022

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computational Science and Engineering (IJCSE):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com