Enabling Proactive Data Management in Virtualized Hadoop Clusters Based on Predicted Data Activity Patterns.
George KousiourisGeorge VafiadisTheodora A. VarvarigouPublished in: 3PGCIC (2013)
Keyphrases
- activity patterns
- data management
- data sets
- database
- data points
- big data
- image data
- data distribution
- data processing
- training data
- data collection
- input data
- synthetic data
- open source
- data mining techniques
- input space
- data records
- data analysis
- data quality
- database management systems
- hierarchical clustering
- data objects
- original data
- raw data
- missing data
- databases
- distributed systems
- data structure