A Class of Communication-avoiding Algorithms for Solving General Dense Linear Systems on CPU/GPU Parallel Machines.
Marc BaboulinSimplice DonfackJack J. DongarraLaura GrigoriAdrien RémyStanimire TomovPublished in: ICCS (2012)
Keyphrases
- parallel machines
- linear systems
- graphics processing units
- parallel computing
- sparse linear systems
- probabilistic model
- scheduling problem
- dynamical systems
- combinatorial optimization
- real time
- genetic algorithm
- computational complexity
- shared memory
- high performance computing
- graphics processors
- preconditioned conjugate gradient method