Research

My research focuses mainly on performance analysis of parallel programs in multi- and many-core architectures. Such architectures include heterogeneous systems featuring GPUs or new processors like the Intel Single-chip Cloud Computer (SCC). Interesting aspects of this analysis include well-known metrics of the application in question such as run time or memory consumption, as well as those that are subject to comparatively new research such as power and energy consumption or reliability and availability considerations.

In the course of my PhD, I am working on the performance measurement and analysis part of Insieme, an optimizing source-to-source compiler that will support multiple parallel frameworks such as OpenMP, MPI and OpenCL.