Umeå University Joins €4.5 Million Consortium to Deliver Context-Aware Cloud Topology Optimisation and Simulation
NEWS
Umeå University today announced it has joined a consortium of leading organisations and universities from the United Kingdom, Ireland and Germany focused on using real-world data to optimize and make more efficient cloud computing.
Named ‘CACTOS’, the €4.5 million project seeks to address issues arising from the growth and complexity of cloud infrastructure. Umeå University will continue its leading research on autonomic cloud management tools. Based on the real cloud usage data gathered by our industrial and academic partners, Umeå University will establish performance models for the heterogeneous landscape of cloud architectures and applications.
These models will be the basis for a set of tools for cloud capacity management. Focus will be on elasticity - determining the optimal amount of resources to allocate, and on scheduling - deciding where in a data centre to allocate those resources. Umeå University will also contribute to construction of the CactoSim, the CACTOS simulator toolkit.
Johan Tordsson, Assistant Professor, said: “CACTOS presents a unique opportunity for us to gain deeper insight into cloud applications and workloads. By understanding these better, we can improve our algorithms and tools for cloud application and infrastructure management. We are also very excited by the strong focus on simulation in CACTOS, which will allow us to experiment and evaluate our methods in very large-scale infrastructures without having access to large data centres.”
About CACTOS
CACTOS is a collaborative research project co-funded under the ICT theme of the 7th framework programme (FP7) of the European Union. For more information visit www.cactosfp7.eu.
About Umeå University
The Distributed Systems Research Group at the Department of Computing Science has extensive experience in research and development of standards-based grid and cloud infrastructure components for job and resource management (including resource brokering, advanced resource selection algorithms, advance reservations, SLAs, elastic auto-scaling of virtual capacity, virtualization in grids, and middleware interoperability); workflow systems (workflow modelling, interoperability, languages, computation models, and applications); and accounting, resource allocation and system management tools. See more at: www.cloudresearch.org
