As more users begin to use clouds for deploying complex applications and store remote data there is an increasing need of minimizing the user costs. In addition many cloud vendors start offering specialized services and thus the need of selecting the best possible service in terms of deadline time or monetary con- straints emerges. Vendors not only that provide specialized services but also prefer using their own scheduling policies and often choose their negotiation strategies. To make things even more complicated complex applications are usually comprised of smaller tasks (e.g. workflow applications) orchestrated together by a composition engine. In this highly dynamic and unpredictable environment multi-agent systems seem to provide one of the best solutions. Agents are by default independent as they act in their best interest following their own policies, but also cooperate with each other in order to achieve a common goal. In the frame of workflow schedul- ing the goal is represented by the minimization of the overall user cost. This paper presents some of the challenges met bymulti-agent systemswhen trying to schedule tasks. Solutions to these issues are also given and a prototype multi agent scheduling platform is presented.
CITATION STYLE
Frîncu, M. E. (2010). Scheduling Service Oriented Workflows Inside Clouds Using an Adaptive Agent Based Approach. In Handbook of Cloud Computing (pp. 159–182). Springer US. https://doi.org/10.1007/978-1-4419-6524-0_7
Mendeley helps you to discover research relevant for your work.