News

Often scientific simulations require more computational resources than locally available. While grid computing already offered on-demand access to large-scale distributed computing resources in the past, cloud computing is much more flexible since it offers the possibility to deploy the full hard- and software stack as desired. However, this increased flexibility also places an additional burden on scientists, who are willing to migrate their simulation applications to the cloud, since it requires a deep understanding of the cloud infrastructure and the related technologies. Therefore, our focus in scope of the project “A cloud-based software infrastructure for distributed simulation applications” is to identify the obstacles scientists face when moving their simulation applications to the cloud and develop a framework to simplify this process. The developed solution is based on the Topology and Orchestration Specification for Cloud Applications (TOSCA) and leverages domain-model knowledge of the scientist to scale the deployed simulation infrastructure.
Our group member Fabian Glaser will talk about results of this project in the context of the SWZ lecture series on Wednesday, March 9th 2016, Room 0.101, Institute for Computer Science, University of Göttingen, 3:30 pm, to which we kindly invite.