Efficient ELM-Based Two Stages Query Processing Optimization for Big Data

Table 2

Feature parameters in the experiments.

Property name	Type	Default value
io.sort.mb	int	100
io.sort.factor	int	10
min.num.spills.for.combine	int	3
mapred.compress.map.output	boolean	false
mapred.reduce.parallel.copies	int	5
mapred.reduce.copy.backoff	int	300
dfs.heartbeat.interval	int	3
dfs.block.size (M)	int	64
mapred.map.task	int	4
mapred.reduce.task	int	4
mapred.tasktracker.map.task.maximum	int	4
mapred.tasktracker.reduce.task.maximum	int	4
Data size (G)	int	10
Data distribution	boolean	uniform
Number of slave nods	int	8