Skip to main content

Table 4 SSB execution times (in seconds): bucketing by “Orderkey” (star schema with buckets (SS-B), denormalized table with buckets (DT-B))

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

  SF = 30 SF = 100 SF = 30 SF = 100
HIVE PRESTO
SS SS-B SS SS-B SS SS-B SS SS-B
Q1.1 25 23 31 29 5 7 13 14
Q1.2 24 23 29 30 5 7 13 13
Q1.3 24 23 29 30 4 6 13 13
Q2.1 32 33 47 59 8 11 19 26
Q2.2 31 32 46 51 7 11 18 23
Q2.3 30 30 44 54 7 10 17 22
Q3.1 35 35 59 64 8 12 29 30
Q3.2 30 30 45 46 6 8 17 19
Q3.3 33 34 219 224 5 8 15 18
Q3.4 34 32 222 225 6 7 15 18
Q4.1 38 39 86 100 13 19 43 47
Q4.2 49 50 70 70 9 14 26 33
Q4.3 34 35 54 65 8 13 23 29
Total 420 421 982 1047 92 133 262 305
Diff.   0%   7%   44%   16%
  SF = 30 SF = 100 SF = 30 SF = 100
HIVE PRESTO
DT DT-B DT DT-B DT DT-B DT DT-B
Q1.1 24 23 29 31 5 5 13 15
Q1.2 24 23 29 30 5 6 14 15
Q1.3 23 23 30 30 5 5 14 15
Q2.1 25 26 36 42 4 6 10 14
Q2.2 36 35 73 69 4 5 10 12
Q2.3 25 23 35 34 4 4 10 10
Q3.1 28 27 40 45 5 5 12 13
Q3.2 28 27 41 44 5 5 12 12
Q3.3 25 24 38 35 4 4 9 10
Q3.4 25 25 38 42 5 5 12 12
Q4.1 27 27 41 44 6 7 14 16
Q4.2 29 29 42 46 6 6 14 17
Q4.3 29 29 42 47 5 6 12 17
Total 349 342 516 539 63 71 155 178
Diff.   − 2%   5%   14%   15%