A time series is a dataset that describe the evolution of a metric overtime, each datapoint contains a value (or several values) and a timestamp.
The time series that we consider in our analysis are downsampled to one minute so each datapoint that streams in is characterised by:
columns | description |
|---|---|
time | timestamp (per minute) in epoch |
v | sum of the values in a minute |
c | number of raw data points in a minute (before downsampling) |
mx | max value within a minute |
environmentId | GUID of the environment |
systemId | system number |
nodeId | node number |
The last three fields help to uniquely identify the time series in the EYER database.