OMB-UM: Design, Implementation, and Evaluation of CUDA Unified Memory Aware MPI Benchmarks.
Karthik Vadambacheri ManianChing-Hsiang ChuAmmar Ahmad AwanKawthar Shafie KhorassaniHari SubramoniPublished in: PMBS@SC (2019)
Keyphrases
- parallel implementation
- general purpose
- efficient implementation
- design process
- design methodology
- case study
- neural network
- parallel distributed
- software development
- implementation issues
- pilot testing
- hardware architecture
- architectural design
- design space
- low power
- design principles
- data structure
- database systems