Skip to main content

Table 5 Clustering (17.79 GB) raster image using a 50 nodes Hadoop-cluster

From: Multi-dimensional geospatial data mining in a distributed environment using MapReduce

Block size (MB)

Elapsed time (min)

Average map time (min)

Average shuffle time (min)

Average merge time (min)

Average reduce time (min)

Total mapping time (min)

128

36.0

13.7

17.9

0.3

1.6

22.9

265

34.5

13.0

14.1

0.3

2.0

21.4

512

36.4

13.7

14.8

0.2

1.7

23.9

1024

49.5

23.5

32.7

0.5

1.9

34.6