Home |  DEUTSCH |  Contact |  Imprint |  Data Protection |  Login |  KIT

DAAD PPP Australia (Sydney)/en

Aus Aifbportal

Wechseln zu: Navigation, Suche
DAAD Logo.svg.png

Automated Runtime & Performance Management of Processes for Domain Experts


Contact: Christian Janiesch




Project Status: completed


Description

In previous and ongoing research, we devised an approach and prototype for monitoring processes and reacting to pre-defined metrics in real-time based on complex event processing. We have evaluated a number of scenarios concerning the usefulness of real-time process feedback and interaction. Enabling domain experts to create their own processes, from design to execution, is an example where real-time feedback is important as the executable process commonly is a black box to the domain expert: Understanding if processes are failing and how to handle exceptions where the processes are created by domain experts is an open problem. In phase 1 of the project, this problem of runtime management will be addressed, i.e. how to best enable domain experts to monitor and manage running processes. Processes are executed on computational resources that could be sourced from commercial Cloud providers. Depending on changes in workload, additional resources may need to be provided to computationally-intensive processes such as scientific data analysis. Therefore, phase 2 of the project will focus on performance management, i.e., how to (near )optimally distribute load on Cloud resources. In phase 3, we will conduct and evaluate case studies with domain experts using a prototypical implementation of our approach in two countries: Australia and Germany to evaluate the design, gather data on usage and effectiveness of our methods, and improve our prototypes. Long-term goals of the collaboration include requesting further joint research projects, such as an EU FP8 project and individual grants through ARC/ DFG. With the proposed project we aim to a) Support domain experts in specifying, executing, and monitoring idiosyncratic processes which assist their daily work by providing design and implementation of software, b) Use process and resource metrics to control the amount of resources available to processes, and c) Conduct and evaluate case studies with domain experts using a prototypical implementation of our design to understand the requirements and needs of domain experts.


Involved Persons
Michael Menzel, Christian Janiesch


Information

from: 1 Januar 2011
until: 31 Dezember 2013
Funding: DAAD


Partners

UNSWNICTA Sydney


Research Group

Economics and Technology of eOrganizations


Area of Research

Business Activity Management


Publications Belonging to the Project
article
 - book
 - incollection
 - booklet
 - proceedings
 - phdthesis
 - techreport
 - deliverable
 - manual
 - misc
 - unpublished





inproceedings
Sven Euting, Christian Janiesch, Robin Fischer, Stefan Tai, Ingo Weber
Scalable Business Process Execution in the Cloud
In IEEE, Proceedings of the 2nd IEEE Conference on Cloud Engineering (IC2E), pages: 175-184, IEEE, Boston, MA, März, 2014
(Details)


Christian Janiesch, Ingo Weber, Michael Menzel, Jörn Kuhlenkamp
Optimizing the Performance of Automated Business Processes Executed on Virtualized Infrastructure
In IEEE, Proceedings of the 47th Hawai'i International Conference on System Sciences (HICSS), pages: 3818-3826, IEEE, Waikoloa, HI, Januar, 2014
(Details)


↑ top