The Umeå University distributed systems’ research focuses flexible and scalable IT infrastructures, and covers a range of topics central to cloud and grid computing.
Research drivers are compute and data intensive applications requiring, e.g., elastic locality-aware infrastructures to meet the rapid capacity and locality variations of industrial services and large-scale distributed environments that enable coordinated use of federated resources for eScience.
Research outcomes include autonomous infrastructure management systems and sophisticated tools for creating cloud-enabled applications. Examples of recent results include algorithms for Virtual Machine (VM) scheduling in clouds, methods for improved live migration of VMs, and algorithms for cloud elasticity control. Further results include tools and principles for (energy) monitoring and accounting, as well as fair-share scheduling systems.
Ongoing projects with immediate industrial benefit include collaboration with Hawc International AB with focus on transforming Hawc’s platform for feature film editing into a cloud service. Another result with industrial application is the creation of the Elastisys spinoff company, with focus on cloud auto-scaling and a product based on the elasticity research.