Experience
    Education
    Bio
    My research interests span distributed and parallel systems, operating systems and networks. I am particularly focused on the performance debugging of parallel and concurrent software and I would love to build the tool that automatically answers the perennial question "Why did my program slow down when run with more cores or more computers?" in a way that is meaningful to the programmer in terms of their algorithm, workload or system configuration. I have worked on performance analysis and characterization for software running on large-scale distributed systems as well as on multicore machines. The three most notable projects that I have been involved in are: Naiad: a distributed, data-parallel computation platform, which was released as open source and is described in our paper at SOSP'13 (best paper award). Naiad supports graph, batch and streaming computations and achieves excellent performance and scaling, even though written entirely in managed code. See http://research.microsoft.com/en-us/projects/naiad/. Barrelfish: a new operating system for future multicore computers that is structured like a distributed system. See the paper at SOSP'11 and download the source at http://www.barrelfish.org/. Magpie: workload characterization for a request-oriented distributed system that captures the resource usage and control path of individual requests. More details at http://research.microsoft.com/en-us/projects/magpie/ and in our OSDI'04 paper. My full list of publications can be found at my Google Scholar page. In addition to my technical work, I currently serve as a member of the EuroSys Steering Committee, and I chair the committee for the Jochen Liedtke Young Researcher Award.