gsmiller commented on pull request #127:
URL: https://github.com/apache/lucene/pull/127#issuecomment-832970411
I also ran a luceneutil benchmark (-source wikimedium10m) to check that the
existing single-valued cases didn't regress in performance. Looks reasonable to
my eye:
```
TaskQPS baseline StdDevQPS range_facet_tweaks
StdDev Pct diff p-value
OrHighNotLow 762.13 (5.9%) 744.85 (6.0%)
-2.3% ( -13% - 10%) 0.231
LowTerm 1149.90 (3.5%) 1136.75 (2.7%)
-1.1% ( -7% - 5%) 0.254
PKLookup 149.74 (2.5%) 148.10 (2.8%)
-1.1% ( -6% - 4%) 0.191
HighTerm 1277.63 (3.7%) 1263.61 (3.2%)
-1.1% ( -7% - 5%) 0.312
Fuzzy2 28.99 (7.1%) 28.68 (6.6%)
-1.0% ( -13% - 13%) 0.633
OrHighLow 325.96 (4.5%) 323.50 (4.1%)
-0.8% ( -8% - 8%) 0.581
Respell 69.99 (1.8%) 69.64 (1.8%)
-0.5% ( -4% - 3%) 0.383
OrNotHighLow 726.65 (2.8%) 723.62 (3.6%)
-0.4% ( -6% - 6%) 0.686
AndHighLow 516.40 (3.1%) 514.33 (2.5%)
-0.4% ( -5% - 5%) 0.655
MedSloppyPhrase 94.03 (3.0%) 93.67 (3.7%)
-0.4% ( -6% - 6%) 0.724
MedTerm 1071.48 (2.8%) 1067.94 (3.0%)
-0.3% ( -5% - 5%) 0.719
Wildcard 74.23 (3.3%) 74.01 (3.5%)
-0.3% ( -6% - 6%) 0.785
MedPhrase 305.87 (3.1%) 305.02 (3.5%)
-0.3% ( -6% - 6%) 0.792
BrowseMonthTaxoFacets 4.88 (3.6%) 4.87 (3.6%)
-0.2% ( -7% - 7%) 0.845
OrNotHighHigh 525.04 (4.2%) 523.90 (4.8%)
-0.2% ( -8% - 9%) 0.878
HighIntervalsOrdered 23.56 (2.0%) 23.53 (1.8%)
-0.1% ( -3% - 3%) 0.813
HighSloppyPhrase 24.69 (4.5%) 24.66 (5.1%)
-0.1% ( -9% - 9%) 0.936
BrowseDateTaxoFacets 4.12 (3.8%) 4.12 (3.9%)
-0.1% ( -7% - 7%) 0.928
OrHighNotHigh 489.22 (4.6%) 488.71 (4.4%)
-0.1% ( -8% - 9%) 0.942
BrowseDayOfYearTaxoFacets 4.12 (3.7%) 4.12 (3.9%)
-0.1% ( -7% - 7%) 0.956
AndHighHigh 50.27 (3.4%) 50.24 (3.1%)
-0.1% ( -6% - 6%) 0.959
OrHighNotMed 554.73 (3.9%) 554.58 (4.4%)
-0.0% ( -7% - 8%) 0.984
LowPhrase 94.40 (2.0%) 94.40 (1.8%)
0.0% ( -3% - 3%) 1.000
IntNRQ 65.05 (0.7%) 65.06 (0.4%)
0.0% ( -1% - 1%) 0.911
AndHighMed 139.73 (3.5%) 139.76 (3.4%)
0.0% ( -6% - 7%) 0.981
LowSloppyPhrase 47.74 (2.8%) 47.77 (3.4%)
0.1% ( -5% - 6%) 0.951
Prefix3 180.61 (3.5%) 180.88 (3.9%)
0.1% ( -7% - 7%) 0.899
HighPhrase 70.08 (2.8%) 70.23 (2.1%)
0.2% ( -4% - 5%) 0.778
LowSpanNear 129.15 (2.5%) 129.47 (3.0%)
0.2% ( -5% - 5%) 0.780
Fuzzy1 61.26 (13.7%) 61.47 (13.6%)
0.3% ( -23% - 32%) 0.937
OrNotHighMed 469.03 (3.8%) 470.84 (3.6%)
0.4% ( -6% - 8%) 0.742
BrowseDayOfYearSSDVFacets 11.06 (5.5%) 11.11 (5.1%)
0.4% ( -9% - 11%) 0.807
MedSpanNear 15.05 (2.0%) 15.12 (2.0%)
0.5% ( -3% - 4%) 0.479
OrHighHigh 19.61 (4.2%) 19.72 (2.3%)
0.5% ( -5% - 7%) 0.624
HighSpanNear 21.58 (2.4%) 21.70 (2.5%)
0.6% ( -4% - 5%) 0.467
OrHighMed 90.57 (3.7%) 91.19 (2.7%)
0.7% ( -5% - 7%) 0.502
BrowseMonthSSDVFacets 12.31 (7.2%) 12.42 (5.9%)
0.9% ( -11% - 15%) 0.682
TermDTSort 100.38 (13.4%) 101.43 (14.1%)
1.0% ( -23% - 32%) 0.810
HighTermMonthSort 128.35 (9.7%) 130.05 (11.9%)
1.3% ( -18% - 25%) 0.700
HighTermTitleBDVSort 121.08 (12.3%) 123.32 (10.9%)
1.8% ( -19% - 28%) 0.616
HighTermDayOfYearSort 92.38 (10.2%) 95.79 (12.7%)
3.7% ( -17% - 29%) 0.312
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
