feat: TraceQL metrics: average over time #4073
Conversation
| default: | ||
| // Simple addition aggregator. It adds existing values with the new sample. | ||
| f = func(existingValue float64, newValue float64) float64 { return existingValue + newValue } | ||
| f = func(b *SimpleAggregator, promLabel string, pos int, newValue float64) { | ||
| b.ss[promLabel].Values[pos] += newValue |
There was a problem hiding this comment.
I think we could also do a Kahan sum here
| @@ -1175,12 +1250,21 @@ func (b *SimpleAggregator) Combine(in []*tempopb.TimeSeries) { | |||
| } | |||
|
|
|||
| b.ss[ts.PromLabels] = existing | |||
| if b.initAvg { | |||
There was a problem hiding this comment.
Struggling a bit with how to include the average in the Combiner. I think this way we allocate the extra bytes only for this case
| } | ||
|
|
||
| func NewSimpleCombiner(req *tempopb.QueryRangeRequest, op SimpleAggregationOp) *SimpleAggregator { | ||
| l := IntervalCount(req.Start, req.End, req.Step) | ||
| var initWithNaN bool | ||
| var f func(existingValue float64, newValue float64) float64 | ||
| var f func(b *SimpleAggregator, promLabel string, pos int, newValue float64) |
There was a problem hiding this comment.
Since we need to modify more than one value on each iteration (counter and compensation) this does the trick
javiermolinar
requested review from
knylander-grafana,
joe-elliott,
annanay25,
mdisibio,
mapno,
kvrhdn,
zalegrala,
electron0zero,
ie-pham and
stoewer
as code owners
|
Not sure if I'm reading this right, but it looks like it's computing the average at each level. I was expecting |
What this PR does:
It implements the avg_over_time aggregation function
Checklist
CHANGELOG.mdupdated - the order of entries should be[CHANGE],[FEATURE],[ENHANCEMENT],[BUGFIX]