Point types

For points collected by interfaces, use the point type that most closely matches the point type in the source system.

For accumulators and other high precision values, use the higher precision point types: Float32 or Float64.

The higher precision point types require more disk space for each stored value. Float16 points use 16 bits or 2 bytes per value. Float32 and Float64 use 4 and 8 bytes per value, respectively. Int16 and Int32 values use 2 and 4 bytes, respectively. Int16 supports numbers from 0 to 32,767.

To store negative integers, select the Int32 point type.

We recommend using Float32 as the default type for floating point data. Float16 can reduce the amount of I/O, but this is significant only on very large data retrievals, such as yearly average calculations or long-term trends.

For more information, see PointType.