The Scientific Data Group constructs enabling technologies for the extraction of knowledge from massive scientific data.
To do this, we focus on techniques for managing, visualizing and analyzing large-scale data from simulations and from experiments. Our approach is based on deep partnerships with leading science teams; this immersion allows us to both push the science forward and have them push us into new innovation. These efforts are disseminated in high-quality venues (journals and conferences) as well as through workforce development, tools, and engagement in support of the ASCR mission.
Our approach drives leadership in research, development, and production deployment across a variety of core topics:
- Information and Data Management: flexible software effectively utilizing the memory hierarchy for on-node, off-node, and over-the-network transport for simulation- and experimental-generated data.
- Extreme Scale Data Services: algorithms, software, and tools for integrative analysis and visualization that are designed for heterogeneous architectures and in situ deployment for our partnership codes.
- Application Experience: deep partnerships with science teams drive the research and development portfolio, and we frequently contribute to science domain papers and end products.
- Community Leadership: We are responsible for contributions to a variety of tools (such as ADIOS, VTK-m, pbdR, VisIt), conferences and workshops (such as SC, IPDPS, LDAV, ISAV, DRBSD, and HPDC), as well as collaborations with industry and academia.
Interested in joining us? See our current permanent and summer student opportunities.