NIAC seeks to advance the use of computing in discovery and a broad range of application areas with a primary focus on:

images-04 Advanced and Future Computing Systems:
Major changes in the computing landscape continue to create opportunities for NIAC to spearhead significant innovation in designing advanced computing systems—integrated co-design that stretches from computer architecture to parallel algorithms. Parallelism on a massive scale impacts every application. Energy efficiency also is a paramount concern and must be addressed by all levels of the system stack. Data-driven discovery—enabled both by the proliferation of sensors and data production of simulations—requires different architectures and algorithms than computational modeling and simulation. Massive supercomputers and cloud architectures present both challenges and opportunities for closely coupled computations.
images-03 Scalable Modeling, Simulation and Design:
Computational modeling and simulation of engineered and natural systems and phenomena form a critical aspect of contemporary engineering and science. In computational modeling and simulation, scientists develop a mathematical model of the phenomena of interest, e.g., the chemical and physical processes involved in an internal combustion engine or the physical, chemical and biological processes involved in the simulation of the climate, and use computers to solve the resulting equations, simulating the system or phenomena of interest. For most systems and phenomena of interest, the equations are complex. Meanwhile, as computing systems extend from today’s petaflops to next-generation exascale, accurate simulations for biological, chemical, physical and, even, social phenomena have advanced dramatically. NIAC will stand at the forefront of scalable techniques on supercomputers, cloud computers and distributed computing systems for design, simulation and modeling.
images-09 Data-driven Science and Discovery:
In data-driven discovery, scientists gather information from data-intensive sources, e.g., large, digitally enabled telescopes; arrays of environmental sensors or genome sequencers; or large-scale simulations, and analyze this mass of data using sophisticated mathematical procedures seeking patterns, information and understanding. Data-driven discovery requires an extensive cyber infrastructure that supports data collection, transport to storage sites and integration and analysis (including visualization). Depending on the quantities of data involved and the mathematical demands for analyses, data-driven discovery may require extensive computing resources and large data storage facilities. As sources of sensed data become more widespread, NIAC will seek to unite these opportunities for linking advanced and future computing systems; scalable modeling, simulation and design; and data-driven science and discovery.