Scientific data management at the Johns Hopkins institute for data intensive engineering and science

Abstract

Scientific computing has long been one of the deep and challenging applications of computer science and data management, from early endeavors in numerical simulation, to recent undertakings in the life sciences, such as genome assembly. Complex computational problems abound and their solutions transform our understanding of the physical world. The data management community’s interest in scientific applications has grown over the last decade due to the commoditization of parallelism, diminishing system administration costs, and a search for relevance beyond enterprise applications. Research in scientific computing raises non-technical challenges, such as overcoming the paucity of resources needed for experimentation, and establishing a collaborative research agenda that fosters a mutual appreciation of problems, results in a concerted effort to develop software tools, and makes all researchers successful in their respective fields. In light of this, we report on a recently formed institute at the Johns Hopkins University to further the interaction between computer science, and science and engineering. We describe ongoing projects at the institute and our collaboration experiences.