You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When viewing large amounts of data for diagnostic purposes, it is often useful to look at derived statistics. These include most notably the Empirical Distribution Function and the Probably Density Function (similar to Histogram). These plots require a little math to generate, but it is all well-documented and available in various third-party libraries: see https://www.npmjs.com/package/distributions
These plots use the viewing time-range as a parameter separate from the 2 dimensions of the display, but provide valuable information about the character of the timeseries data. This feature request is related to #600#516#949
The text was updated successfully, but these errors were encountered:
graphite has a bunch of analytics functions, if your time series store is missing a metric transform function when please open a PR for that time series store :)
Grafana leave's the processing of metric queries and metric transformation functions to the time series backends. Processing metric queries and doing transformation is much better done there.
When viewing large amounts of data for diagnostic purposes, it is often useful to look at derived statistics. These include most notably the Empirical Distribution Function and the Probably Density Function (similar to Histogram). These plots require a little math to generate, but it is all well-documented and available in various third-party libraries: see https://www.npmjs.com/package/distributions
These plots use the viewing time-range as a parameter separate from the 2 dimensions of the display, but provide valuable information about the character of the timeseries data. This feature request is related to #600 #516 #949
The text was updated successfully, but these errors were encountered: