Beginning with the 2018-19 incoming class, the DOE CSGF now accepts doctoral students engaged in traditional, application-focused computational science research and those pursuing applied mathematics, statistics or computer science research enabling the use of emerging high-performance computing (HPC) systems without a specific application. Together, they will meet the DOE's growing needs to build and efficiently operate the latest, most powerful HPC systems and apply those systems to problems of national importance.

Since 1991, the DOE CSGF has supported doctoral students in fields of study that advance the use of high-performance computing (HPC) to solve specific, complex science and engineering problems. These fields span science and technology, ranging from engineering to biological sciences and from astrophysics to geophysics.

Those who fit this traditional DOE CSGF definition should apply here:
https://www.krellinst.org/doecsgf/application

### New DOE CSGF Math/Computer Science Track

As of 2018, the DOE CSGF has expanded to accept doctoral candidates researching HPC-enabling technologies in applied mathematics, statistics or computer science. Unlike traditional DOE CSGF students, Math/Computer Science fellows need not focus on a particular science or engineering application. You can find additional details and examples of appropriate research areas here.

Those who fit the Math/Computer Science Track description should apply here:
https://www.krellinst.org/doecsgf/application/mathcs.shtml