Targeting at new paradigms for SSIO in the exascale era, we are investigating novel “object-centric” data abstractions and storage mechanisms that take advantage of the deep storage hierarchy. In order to achieve this overarching goal, we have proposed a fundamental new data abstraction, called Proactive Data Containers (PDC). A PDC is a container in a storage location (memory, non-volatile memory, disk, etc.) that stores scientific data in an object-oriented manner. We are researching a formulation of object-oriented PDCs, efficient strategies for moving and reorganizing data using PDCs, and novel analysis paradigms. We are delivering the developed research techniques in the HDF5 library, which is used by numerous scientific applications. We are also exploring new programming interfaces that are suitable for managing data as objects. These proactive data movement strategies, simple object-based interfaces, and analysis methods are aimed at reducing the burden on scientific application developers.
The novel data management and storage paradigms, approaches, and formalisms being developed in this project are expected to be applicable to a broad range of scientific and engineering problems. These science applications use computational and experimental facilities for predictive understanding of physical processes through data analytics and visualization and the proposed techniques are expected to accelerate the crucial process of data-driven exploration and knowledge discovery.