Skip to main content

Advances, Systems and Applications

Table 3 Execution times of sequential and merged queries in seconds

From: Improving the performance of Hadoop Hive by sharing scan and computation tasks

Query names

Correlation

Execution time (sec.) Hive/SharedHive (reduction %)

  

1GB

100GB

1TB

11,12 (set 1)

none

231/215 (6.9%)

676/666 (1.5%)

5,410/5,386 (0.4%)

17,22 (set 2)

none

272/262 (3.7%)

1,382/1.323 (4,3%)

11,078/11,060 (0.2%)

1,17 (set 3)

partial

192/185 (3.6%)

1,330/1,134 (14.7%)

12,300/11,386 (7.4%)

1,18 (set 4)

partial

252/248 (1.6%)

1,540/1,396 (9.4%)

17,499/16,324 (6.7%)

6,17 (set 5)

partial

163/160 (1.8%)

1,169/1,042 (10.9%)

10,430/8,636 (17.2%)

1,6 (set 6)

full

103/90 (12.6%)

601/436 (27.5%)

5,057/3,936 (22.2%)

14,19 (set 7)

full

138/83 (39.9%)

1,004/789 (21.9%)

8,989/7,178 (20.1%)

1,14,18 (set 10)

mixed

317/299 (5.7%)

1,870/1,689 (9.7%)

20,425/18,590 (9.0%)

1,3,11,14,17,19 (set 11)

mixed

620/524 (15.5%)

3,232/2,830 (12.4%)

30,069/26,024 (13.5%)