Skip to main content

Table 13 SSB execution times (in seconds): partitioning by “Od_Year” and “S_Region” and bucketing by “Suppkey”.

From: Evaluating partitioning and bucketing strategies for Hive-based Big Data Warehousing systems

  SF = 30 SF = 100 SF = 30 SF = 100
HIVE PRESTO
DT DT-PB DT DT-PB DT DT-PB DT DT-PB
Q1.1 24 19 29 23 5 2 13 5
Q1.2 24 22 29 44 5 3 14 8
Q1.3 23 18 30 25 5 3 14 6
Q2.1 25 20 36 25 4 2 10 6
Q2.2 36 22 73 35 4 2 10 5
Q2.3 25 19 35 24 4 2 10 4
Q3.1 28 19 40 27 5 2 12 6
Q3.2 28 23 41 47 5 3 12 11
Q3.3 25 23 38 45 4 3 9 12
Q3.4 25 25 38 51 5 4 12 13
Q4.1 27 20 41 28 6 3 14 7
Q4.2 29 15 42 22 6 2 14 3
Q4.3 29 21 42 29 5 2 12 5
Total 349 265 516 424 63 33 155 90
Diff   − 24%   − 18%   − 47%   − 42%
  1. Italic values indicate the fastest processing time by query, workload, tool and data model