Parallel and Distributed Computing
| Research Areas -> Theory of Computation -> Parallel and Distributed Computing |
| Overview |
|
For decades the speed of processors has grown exponentially, making today’s most computation-intensive applications (e.g., video games) possible. However, due to basic physical barriers this exponential growth cannot continue much longer—unless we find a way to exploit the parallelism intrinsic to the problem we want to solve. At Michigan we are investigating the power of different models of parallel computation and developing algorithms and data structures for actual parallel computers. For example, we may have a vast number of tiny processors arranged in a grid, where each can communicate only with its neighbors or, alternatively, to very distant processors via long optical wires. Whether this type of parallel computer can give some speedup over a serial computer (or quantum computer) actually depends on the problem you want to solve! We also study abstract models of distributed computation, where a number of independent, unsynchronized computers are arranged in a (possibly unknown) network and must solve a problem only through local communication. |
|
Faculty Pettie, Seth Stout, Quentin F. Related Labs, Centers, and Groups Software Systems Laboratory |
|