High-throughput analysis of large heterogeneous and dynamic data spaces with signac
ORAL
Abstract
High-throughput generation and analysis of vast data sets offers enormous opportunities for accelerated scientific discovery, but also requires prudent strategies for the management of computational resources and data spaces. This is especially critical when researchers work with heterogeneous and possibly highly dynamic data. The signac framework enables researchers to maintain well-formed and reusable data spaces from early exploration all the way to production runs on supercomputing scales. This is achieved through a transparent data and workflow model as well as a simple and unobtrusive programmatic interface that scales well between preliminary prototyping and concluding stages of a particular computational investigation. Here, we demonstrate the framework's efficacy and versatility by showcasing examples of how signac is applied across various research projects and disciplines.
*Development and deployment supported by UM and MICCoM, as part of the Computational Materials Sciences Program funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division, under Subcontract No. 6F-30844. Project conceptualization and early implementation supported by the National Science Foundation, Award # DMR 1409620.
–
Presenters
-
Carl Simon Adorf
- University of Michigan
- Chemical Engineering, University of Michigan