A Data-Oriented Method for Scheduling Dependent Tasks on High-Density Multi-GPU Systems.
Peng ZhangYu-Xiang GaoMeikang QiuPublished in: HPCC/CSS/ICESS (2015)
Keyphrases
- synthetic data
- high density
- test data
- input data
- data sets
- noisy data
- data structure
- prior knowledge
- database
- computer systems
- close proximity
- statistical methods
- missing data
- data analysis
- data mining techniques
- training samples
- training data
- distributed systems
- data sources
- pairwise
- particle filter
- data streams
- low cost
- support vector machine
- missing values
- scheduling problem
- feature selection
- parallel computing
- graphics processing units
- low density