Metrics Preconfigs¶
Metrics preconfigs should be put in
<autotest_dir>/new_tko/tko/preconfigs/metrics/
The parameters are:
- plot: Line or Bar
- xAxis: Database column name for the X-axis values control. See GraphingDatabaseFields.
- globalFilter[i][db]: Database column name for the ithglobal filter (start at 0). See GraphingDatabaseFields.
- globalFilter[i][condition]: Condition field for the ith global filter (start at 0).
- globalFilter_all: This controls if you have “all of” or “any of”
selected as the filter combination operation for the global filters.
Set to
truefor “all of”, andfalsefor “any of”. - name[j]: The name of the jth series.
- values[j]: The database column name that should be plotted on the y-axis for the jth series. See GraphingDatabaseFields.
- aggregation[j]: The aggregation to be applied to the data of the
jth series. Available aggregations are:
- AVG
- COUNT (DISTINCT)
- MIN
- MAX
- errorBars[j]: Sets if the error bars should be shown for the
jth series, if the aggregation is AVG. Set to
trueto show error bars,falseto keep them hidden. - seriesFilters[j][k][db]: Database column name for the kth filter of the jth series. See GraphingDatabaseFields.
- seriesFilters[j][k][condition]: Condition field for the kth filter of the jth series.
- seriesFilters[j]_all: This controls if you have “all of” or “any
of” selected as the filter combination operation for the filters on
the jth series. Set to
truefor “all of”, andfalsefor “any of”.
Example:
plot: Line
xAxis: kernel
globalFilter[0][db]: hostname
globalFilter[0][condition]: = 'my_test_host'
globalFilter_all: true
name[0]: dbench (throughput)
values[0]: iteration_value
aggregation[0]: AVG
errorBars[0]: true
seriesFilters[0][0][db]: iteration_key
seriesFilters[0][0][condition]: = 'throughput'
seriesFilters[0][1][db]: test_name
seriesFilters[0][1][condition]: = 'dbench'
seriesFilters[0]_all: true
name[1]: unixbench (score)
values[1]: iteration_value
aggregation[1]: AVG
errorBars[1]: true
seriesFilters[1][0][db]: iteration_key
seriesFilters[1][0][condition]: = 'score'
seriesFilters[1][1][db]: test_name
seriesFilters[1][1][condition]: = 'unixbench'
seriesFilters[1]_all: true