As computing power for simulation-based analysis approaches the exascale, the need will arise for practical, efficient, robust, and fault-tolerant exascale data management and querying capabilities. Modern physical simulations produce prodigious amounts of data that must be compared, studied, and tested against experiments, theory, and other simulations. Moreover, simulation results contain uncertainties that typically require an ensemble of runs to characterize; such ensembles result in massive databases. We are interested in methods for querying these massive databases.
This blog is a place to discuss the array of topics related to such queries:
- Computational mathematics and numerical analysis,
- Probabilistic methods and uncertainty quantification,
- High performance scientific computing,
- Machine learning and descriptive statistics,
- Physics models and discretization methods,
- Hardware and software for scientific computing.
Occasionally, we may include brief epistles (you know, rants) on things like
- The academic life,
- Science and engineering in the public sphere,
- Relevant political discourse,
- Whatever else seems pertinent.