Communication lower bounds for distributed-memory matrix multiplication.
Dror IronySivan ToledoAlexandre TiskinPublished in: J. Parallel Distributed Comput. (2004)
Keyphrases
- matrix multiplication
- distributed memory
- lower bound
- shared memory
- parallel implementation
- multithreading
- multiprocessor systems
- upper bound
- ibm sp
- scientific computing
- message passing
- parallel architecture
- objective function
- parallel machines
- communication networks
- three dimensional
- parallel computers
- np hard
- data parallelism
- graph cuts
- optimal solution