Research Article

Application of Filters to Multiway Joins in MapReduce

Table 1

Test datasets for common attribute joins.

Table Number of records Size Number of records satisfying selection predicate

Part 20 M 2.3 GB 1.08 M
Partsupp 80 M 12 GB 80 M * 
Lineitem 600 M 75 GB 600 M (no predicate)