An Architectural Style for Highly Data-Intensive Systems

Poster

Co-Researcher: Daniel J. Crichton, Jet Propulsion Laboratory

Abstract: Data-Intensive software systems are large-scale, distributed systems which integrate information from heterogeneous data resources. As data volumes approach near petabyte scales, data-intensive systems are becoming a focus area of research in the distributed system and super-computing communities, yet curiously not in the software engineering communities. In particular, lack of design and implementation-level reuse in data-intensive software systems has yielded a distasteful practice of re-implementing software systems to accompany data collection, distribution and processing tasks at research institutions, universities and large corporations throughout the world; rather than leveraging existing frameworks, code and tools. The focus of my research is applying the practice software architecture to address these important issues. My initial work in this area has resulted in the formulation of the OODT architectural style for data-intensive systems. OODT was initially developed at NASA’s Jet Propulsion Laboratory in the form of a middleware infrastructure for earth science data systems. The current work ongoing between JPL and USC (and the focus of my research) is formalizing OODT and extracting its architectural style. To date, the OODT architectural style has been used to design and implement large-scale information integration systems at both NASA Planetary Data System and the National Cancer Institute’s Early Detection Research Network – two highly heterogeneous, yet similar scientific domains.

Bio: Chris A. Mattmann is a Ph.D. student studying Computer Science at the University of Southern California (USC). At USC, Chris is a member of the Software Architecture Research Group, which is led by his advisor, Dr. Nenad Medvidovic. Chris also works at NASA’s Jet Propulsion Laboratory, where he is an Associate Software Engineer in the Earth Science Data Systems Section's OODT Group, led by his project manager Daniel J. Crichton. Chris received the B.S. degree in Computer Science and the M.S. degree in Computer Science with a Specialization in Multimedia and Creative Technologies, from the Department of Computer Science at USC in 2001 and 2003 respectively. Chris’s research interests lie in Software Architecture for Data Intensive Systems, Architectural Styles of Data Grids, Grid Computing, and Distributed Software Architectures. He is a member of IEEE, ACM SIGSOFT, and participates in CCSDS (Consultative Committee on Space Data Systems).