What Is Aggregation?

Aggregation is a process where Cloud Eye aggregates the maximum, minimum, average, sum, or variance value of raw data sampled for different periods, and this process repeats for each subsequent period. Each period is called an aggregation period. A calculation period is called an aggregation period.

During the aggregation, data sets are smoothed. A longer aggregation period means more smoothing and precise data, enabling you to predict trends more precisely. On the contrary, a shorter aggregation period means more accurate alarms.

The aggregation period can be 5 minutes, 20 minutes, 1 hour, 4 hours, or 1 day.

During the aggregation, Cloud Eye processes the sampled data based on the data type.

• If the data sampled is integers, Cloud Eye rounds off the aggregation results.
• If the data includes decimal values (floating point number), Cloud Eye truncates the data after the second decimal place.

For example, the instance quantity in Auto Scaling is an integer. If the aggregation period is 5 minutes, and the current time is 10:35, Cloud Eye will aggregate the raw data generated between 10:30 and 10:35 to the time point of 10:30. If the raw data 1 and 4, after aggregation, the maximum value is 4, the minimum value is 1, and the average value is [(1 + 4)/2] = 2, instead of 2.5.

Choose whichever aggregation method best meets your service requirements.