Argonne’s Mathematics and Computer Science division researches new techniques and tools for storing, transferring, accessing, and manipulating extremely large data. The goal of this work is to enable researchers to analyze, visualize, and query these large datasets to lead to new science discoveries and breakthroughs.
Addressing the challenges raised by the management of the massive datasets produced by simulation and advanced instruments, through projects such as EZ: Fast, Effective, Parallel Error-Bounded Exascale Lossy Compression for Scientific Data and Data Libraries and Services Enabling Exascale Science.
Enabling the rapid transfer of the vast amounts of data being produced by large-scale scientific simulations and advanced instruments, through projects such as Globus Online, Improving Data Movement on Large HPC Systems, DIY, and Exploring Efficient Data Movement Strategies for Exascale Systems with Deep Memory Hierarchies.
Providing solutions to the problem of big data storage, through project such as Data Model Storage Library for Exascale Science, Enabling Exascale Science through Node Local Storage Systems, Enabling Co-Design of Multilayer Exascale Storage Architectures, and DAOS.