I am an Associate Professor in the Department of Computer Science at Duke University. My primary research interest is in making data-intensive computing systems easier to manage. Recent work from my research group has focused on the Hadoop MapReduce system. Check out our Starfish project. I am very interested in using cloud platforms for online experimentation to aid system tuning and testing. The vision of our Flex project is to enable users---irrespective of whether they are end-users, developers, or system administrators---to have programmatic access to collect information needed for system testing and tuning through planned experiments on the cloud. The DIADS project tackles integrated problem diagnosis for database systems running on networked storage as well as automated detection and recovery from data corruption caused by hardware faults, software bugs, or human mistakes. Our work is supported by startup funds from Duke, grants from the US National Science Foundation, faculty awards from IBM, an equipment grant from IBM, and resource usage grants from Amazon Web Services.