The doctoral program in scientific computing was established by the University of Michigan in 1989. This program is intended for students who will make extensive use of large-scale computation, computational methods, or algorithms for advanced computer architectures in their doctoral studies. A firm knowledge of the scientific discipline is essential.

This is not a stand-alone degree; this is a joint degree program. Students must be accepted into the PhD program of a home department. The actual degree name will have “…and Scientific Computing” appended to the the normal title, e.g., “Ph.D. Degree in Aerospace Engineering and Scientific Computing.”

Students may apply to the program after having completed one term, but prior to being promoted to candidacy status.

Graduation Requirements

Students are expected to complete the normal doctoral requirements of their home department as well as meet additional requirements in the area of scientific computation. The specific requirements will vary with the choice of home department; however, there are a few general course requirements for all students in this program. The student must elect:

• nine (9) credit hours in numerical analysis, and
• nine (9) credit hours in computer science and applications in scientific computation outside the home department.

The Center for Advanced Computing

The Center for Advanced Computing (CAC) administers the doctoral degree program in scientific computing, in cooperation with the student’s home department. The CAC was created in 2002 in response to the increased emphasis on scientific computation in nearly all applications areas, and is intended to foster and encourage research and instructional activities in scientific computation at the university.

The CAC administers more than 4000 nodes of cluster-based systems, in conjunction with the College of Engineering Computer-Aided Engineering Network (CAEN). CAC maintains close ties with the National Science Foundation’s supercomputer centers at San Diego and Cornell, providing high speed network links to the many vector supercomputers and parallel machines at those sites.

The CAC staff consists of four permanent staff members. They maintain the existing high-performance computing systems and support faculty who wish to buy equipment or need dedicated access.

Fellowship in Scientific Computing

U-M is an approved institution for the Department of Energy (DOE) Computational Science Graduate Fellowship Program. This program is designed to support outstanding science and engineering students interested in pursuing doctoral study in applied science or engineering disciplines with applications in high-performance computing. This four-year renewable fellowship provides:

• a $36,000 yearly stipend
• payment of all tuition and fees
• yearly conferences
• $5,000 academic allowance in the first fellowship year
• $1,000 academic allowance each renewed year
• 12-week research practicum

This program is sponsored is administered by the Krell Institute and the CAC. For more information, visit krellinst.org/csgf.

Students should apply to their home departments for other financial assistance, including research or teaching assistantships.

Areas of Research