New Search Based Methods to Solve Workflow Scheduling Problem in Cloud Computing

Scheduling has a big influence on the performance of the cloud computing environment, and still remains of big interest for researchers, in particular when a certain level of quality should be maintained in order to satisfy the customer. We consider the particular scheduling multiobjective optimizat...

Full description

Saved in:
Bibliographic Details
Main Authors: Bouzidi, Mohammed Ridha, Soltani, Abdelghani, Bouhank, Asma, Daoudi, Mourad
Format: Conference Proceeding
Language:English
Subjects:
Cloud computing
Measurement
Processor scheduling
Quality of service
Reliability
Task analysis
Virtual machining
Citations: Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Scheduling has a big influence on the performance of the cloud computing environment, and still remains of big interest for researchers, in particular when a certain level of quality should be maintained in order to satisfy the customer. We consider the particular scheduling multiobjective optimization problem of allocating workflows to the resources of a cloud computing environment, by managing four QoS metrics: makespan, cost, reliability and the availability. It belongs to a category of NP Hard problems. A particular metaheuristic, BBO is investigated. New workflow scheduling BBO based methods are proposed. Further, two multiobjective pareto based optimization methods MOHEFT and NSGA-II are considered in solving our problem. Tests are performed on well-known benchmarks, showing a good behavior of the different methods.
ISSN:2576-3555
2576-3555
DOI:10.1109/CoDIT.2018.8394855