Tight Memory-Independent Parallel Matrix Multiplication Communication Lower Bounds.
Hussam Al DaasGrey BallardLaura GrigoriSuraj KumarKathryn RousePublished in: CoRR (2022)
Keyphrases
- lower bound
- matrix multiplication
- distributed memory
- upper bound
- multithreading
- shared memory
- branch and bound algorithm
- branch and bound
- worst case
- parallel hardware
- message passing
- parallel implementation
- np hard
- objective function
- lower and upper bounds
- memory requirements
- optimal solution
- parallel processing
- matrix factorization
- parallel machines
- high quality
- three dimensional
- post processing