In January 2019, the U.S. Department of Energy (DOE), Office of Science, Office of Advanced Scientific Computing Research (ASCR) convened a workshop on In Situ Data Management (ISDM). Tom Peterka, a computer scientist in Argonne’s Mathematics and Computer Science Division, chaired the workshop and, with help of an organizing committee drawn from five other DOE national labs, prepared the workshop report.
Why should scientists care about managing data in situ? First, it can benefit data-driven scientific discovery from a wide range of sources, including simulation and experiments. Second, it can help decrease large data volumes, thereby minimizing data movement, saving storage space and increasing resource efficiency.
The workshop identified six priority research directions needed for ISDM to facilitate scientific discovery: (1) enabling ISDM to execute over diverse computational platforms, (2) providing flexible interoperable software components, (3) promoting intelligent design decisions through better understanding of the design space, (4) capturing provenance and metadata for assessing scientific results, (5) coordinating ISDM tool design with the software stack; and (6) developing fundamentally new data algorithms.
For details about the workshop, see the website https://www.orau.gov/insitudata2019/default.htm