Table of Contents

Trending techniques

There are two trending techniques: real-time trending and average trending.

Logging of all values, in a sliding window of maximum 1000 hours.

In most cases, real-time trending will be set to 24 or 48 hours.

Note

It is possible to set a default sliding window size for real-time trending on a particular DMA. See Setting the default sliding window size for real-time trending.

Logging of average values only, usually across longer time spans than those used for real-time trending.

  • Last 48 hours, with 5-minute averages.

  • Last week, with 5-minute averages (max. 2016 data points).

  • Last month, with 5-minute averages (max. about 8640 data points).

  • Last year, with 1-hour averages (max. about 8760 data points).

Note

For average trending, the DMS stores more than just the average value for a particular time slot. For numeric parameters, it also stores the minimum and maximum value detected in that same time slot. For non-numeric parameters (such as discrete and hybrid parameters), the DMS stores the value that occurred the most, the value that was active the longest, and the percentage of time this value was active during the time slot.

Propose changes