Interruptible tasks: treating memory pressure as interrupts for highly scalable data-parallel programs.
Lu FangKhanh NguyenGuoqing XuBrian DemskyShan LuPublished in: SOSP (2015)
Keyphrases
- highly scalable
- data sets
- data analysis
- raw data
- data structure
- data sources
- data transfer
- data processing
- original data
- training data
- application programs
- xml documents
- synthetic data
- data distribution
- data objects
- database
- incoming data
- computational power
- input data
- knowledge discovery
- probability distribution
- data streams
- neural network