Workflow scheduling and resource allocation for cloud-based execution of elastic processes

Publikationstyp
Conference Paper
Date Issued
2013-12
Sprache
English
Author(s)
Hoenisch, Philipp  
Schulte, Stefan  
Dustdar, Schahram  
TORE-URI
http://hdl.handle.net/11420/11943
Start Page
1
End Page
8
Article Number
6717277
Citation
IEEE 6th International Conference on Service-Oriented Computing and Applications (SOCA 2013)
Contribution to Conference
IEEE 6th International Conference on Service-Oriented Computing and Applications, SOCA 2013  
Publisher DOI
10.1109/SOCA.2013.44
Scopus ID
2-s2.0-84894204588
Today's extensive business process landscapes make it necessary to handle the execution of a large number of workflows. Especially if workflow steps require the invocation of resource-intensive applications or a large number of applications needs to be carried out concurrently, process owners may have to allocate extensive computational resources, leading to high fixed costs. Instead, process owners could make use of Cloud-based computational resources, dynamically leasing and releasing resources on demand, which could lead to lower costs. In the work at hand, we propose a resource-efficient workflow scheduling algorithm for business processes and Cloud-based computational resources. Through the integration into the Vienna Platform for Elastic Processes and an evaluation, we show the practical applicability and the benefits of our contributions. Specifically, we find that our approach reduces the resource demand if compared with an ad hoc approach.
Subjects
Business Process Execution
Cloud Computing
Elastic Processes
Workflow Scheduling