Moving algorithms are calculations that use the values in the current sample and values from previous samples to provide historic values. They allow you to set the system to calculate a value based on a time period that is a different size to the interval. For example, you can use a moving algorithm to calculate the average hourly value every 30 minutes. The one hour algorithm sets the duration of the sample, and the interval of 30 minutes is set for the trace on a Trend (or in a Historic View that can be accessed by an SQL Query).
Moving algorithms can only be used to calculate sum or average values.
Example:
An 8 hour algorithm is used to calculate the average values.
The average value for an 8 hour period is to be calculated once an hour.
Every hour, the algorithm calculates an average value based on the values that have been reported during the previous 8 hours.
For example, at 3pm, the algorithm calculates the average value based on the values that have been reported during the hours 7am-8am, 8am-9am, 9am-10am, 10am-11am, 11am-12 noon, 12 noon-1pm, 1pm-2pm, and 2pm-3pm.
At 4pm, the algorithm calculates the average value based on the values that have been reported during the hours 8am-9am, 9am-10am, 10am-11am, 11am-12 noon, 12 noon-1pm, 1pm-2pm, 2pm-3pm, and 3pm-4pm.
This means that for every hour in the day there is an average value based on the previous 8 hours worth of data.
If an accumulative algorithm had been used to calculate the average value for 8 hours of data, the historic value would only be calculated 3 times a day—once every 8 hours.