Handling Data Skew in MapReduce Cluster by Using Partition Tuning

Algorithm 1

Repartitioning algorithm.

 Data:

 Result: R: an index of subsequence

1

2

3

4 while do

5  

6  foreach do

7     

8     if then

9       

10     

11     

12     end

13  end

14  if then

15     

16  end

17  else if then

18     

19  end

20 end

21 return ;