Multiple Fields In Calculations

To use values from multiple fields in a mathematic calculation, complete the following steps:

1. Filter by fields required in your calculation
2. Pivot fields into columns
3. Perform the mathematic calculation

Filter by fields

Use filter() to return only the fields necessary for your calculation. Use the or logical operator to filter by multiple fields.

The following example queries two fields, A and B:

from(bucket: "example-bucket")
    |> range(start: -1m)
    |> filter(fn: (r) => r._field == "A" or r._field == "B")

This query returns one or more tables for each field. For example:

{{< flex >}} {{% flex-content %}}

Pivot fields into columns

Use pivot() to align multiple fields by time.

{{% note %}} To correctly pivot on _time, points for each field must have identical timestamps. If timestamps are irregular or do not align perfectly, see Normalize irregular timestamps. {{% /note %}}

// ...
  |> pivot(rowKey: ["_time"], columnKey: ["_field"], valueColumn: "_value")

Using the queried data above, this pivot() function returns:

Perform the calculation

Use map() to perform the mathematic operation using column values as operands.

The following example uses values in the A and B columns to calculate a new _value column:

// ...
    |> map(fn: (r) => ({ r with _value: r.A * r.B }))

Using the pivoted data above, this map() function returns:

Full example query

from(bucket: "example-bucket")
    |> range(start: -1m)
    |> filter(fn: (r) => r._field == "A" or r._field == "B")
    |> pivot(rowKey: ["_time"], columnKey: ["_field"], valueColumn: "_value")
    |> map(fn: (r) => ({r with _value: r.A * r.B}))