Research Article

Application of Filters to Multiway Joins in MapReduce

Table 2

Test datasets for distinct attribute joins.

Table Number of records Size Number of records satisfying selection predicate

Supplier 3 M 0.4 GB 0.6 M
Part 60 M 6.9 GB 60 M * 0.2 * 
Partsupp 240 M 35 GB 240 M (no predicate)