Checkpointing Kernel Executions of MPI+CUDA Applications.
Max BairdSven-Bodo ScholzArtjoms SinkarovsLeonardo Bautista-GomezPublished in: Euro-Par Workshops (2019)
Keyphrases
- parallel implementation
- general purpose
- parallel computing
- shared memory
- parallel programming
- distributed databases
- kernel function
- kernel methods
- low overhead
- message passing interface
- parallel algorithm
- fault tolerance
- message passing
- support vector
- distributed database systems
- main memory databases
- kernel machines
- kernel regression
- high performance computing
- compute unified device architecture
- feature space
- gpu implementation
- parallelization strategy
- failure recovery
- multiple kernel learning
- kernel pca
- distributed memory
- parallel computation
- reproducing kernel hilbert space
- machine learning
- cloud computing
- feature extraction