Skip to main content

Table 12 SSB execution times (in seconds): partitioning by “Od_Year” and bucketing by “P_Brand” (denormalized table with partitions and buckets (DT-PB)).

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 21 5 2 13 3
Q1.2 24 21 29 22 5 2 14 5
Q1.3 23 20 30 21 5 2 14 4
Q2.1 25 26 36 40 4 5 10 14
Q2.2 36 36 73 68 4 4 10 11
Q2.3 25 23 35 32 4 4 10 9
Q3.1 28 25 40 40 5 5 12 13
Q3.2 28 26 41 39 5 5 12 13
Q3.3 25 22 38 31 4 4 9 10
Q3.4 25 25 38 39 5 4 12 10
Q4.1 27 27 41 43 6 6 14 17
Q4.2 29 22 42 26 6 3 14 5
Q4.3 29 21 42 27 5 3 12 5
Total 349 312 516 449 63 47 155 119
Diff   − 10%   − 13%   − 24%   − 23%
  1. Italic values indicate the fastest processing time by query, workload, tool and data model