Research

The team takes a scientific approach to the research, development, and deployment of algorithms and software that enable computational science. Some of our current research activities include:

Darshan I/O Observation Tool. A key component of our research strategy is to maintain a solid understanding of the needs of our customers: DOE computational scientists. In addition to meeting regularly with science teams, we have developed Darshan, a tool for observing I/O behavior on large scale compute platforms.

Mochi Software Defined Storage. Ultimately, we aspire to create software products that are used by the community. The objective the Mochi project is to explore a software defined storage approach for composing storage services that provides new levels of functionality, performance, and reliability for science applications at extreme scale. The Mochi project is a collaboration between Argonne National Laboratory, Los Alamos National Laboratory, Carnegie Mellon University, and the HDF Group.

CODES Simulation Framework. Because we are developing system software that targets future systems, simulation is another key ingredient of our research strategy. CODES is our framework for parallel discrete event simulation (PDES) of large scale data management systems, being developed with collaborators at Rensselaer Polytechnic Institute (RPI) and others, that builds on the RPI team’s ROSS PDES system.

I primarily provide technical guidance for these projects, with occasional coding when I can find time. More information on current, data related research activities in MCS can be found here.