Assistant with doctoral duties at Department of Information Technology, Division of Computer Systems
My work revolves around task management on heterogeneous architectures. Currently I am investigating how to efficiently utilize GPUs with tasks that does not scale well, either due to limited parallelism, or other more complex reasons. This has so far led me down paths such as performance modelling, concurrent execution, and the starPU runtime system.
Also available at
Please contact the directory administrator for the organization (department or similar) to correct possible errors in the information.