Igor, I think you may mean that the standard deviation is calculated over the past 1 day of data in your example (rather than timestamp).
Thanks for your feedbacks.
In fact, my users would like to calculate standard deviation of the following sample series.
tag1 values: 0.5, 0.6, 0.7, 0.2
Above values are gotten at the time where tag2 turns from 0 to 1.
The calculation could be the discrete stdev of those 4 values as example.
One way I tried yesterday is that create and output tag3 when tag2 has value change and tag2 = 1. In this case, we can have values stored in PI DA Server as 0.5, 0.6, 0.7, 0.2 series, not more repeated values. (Even though I observed some missing value at timestamp where tag2 meets criteria. Don't know why. after backfilling it, the missing one is filled).
However, this only gives the stored series in tag3. Any straight way to get stdev according to the discrete value series stored ?
Thanks for your response.