boxplot

Calculates the minimum, maximum, and quartile values needed to draw a box plot. You can optionally split the statistics by group.

Command properties

Item	Description
Command type	Processing query
Required permission	None
License usage	N/A
Parallel execution	Not supported
Distributed execution	Runs on Control Node (reducer)

Syntax

boxplot EXPR [by FIELD, ...]

Target

EXPR: Expression to compute statistics on. Records where the expression result is null are ignored.
[by FIELD, ...]: Group fields. Separate multiple fields with commas (,). When you specify the by clause, box plot statistics are calculated independently for records that share the same group field values.

Output fields

Field	Type	Description
min	numeric	Minimum value per group
iqr1	numeric	First quartile (Q1) per group: the value at the 25th percentile of all data
iqr2	numeric	Second quartile (Q2, median) per group: the value at the 50th percentile
iqr3	numeric	Third quartile (Q3) per group: the value at the 75th percentile of all data
max	numeric	Maximum value per group
count	integer	Record count per group

Error codes

Parse errors

Error code	Message	Description
20000	Incomplete expression	The syntax is incomplete, for example the `by` clause ends with a comma.
20001	No expression value entered	No expression is specified for the statistics target.

Runtime errors

N/A

Description

The boxplot command collects all input records, sorts them by the value of the specified expression, and calculates the minimum, maximum, and quartile values. When you use the by clause, statistics are calculated independently for records that share the same group field values.

Records where the expression result is null are excluded from the statistics. Output records include the min, iqr1, iqr2, iqr3, max, and count fields. When you specify a by clause, the group fields are also included in the output.

In a distributed environment, the final calculation runs on the Control Node.

Examples

Computing summary statistics for all CPU usage values
```
json "[{'usage': 10}, {'usage': 25}, {'usage': 50}, {'usage': 75}, {'usage': 90}]"
| boxplot usage
```
Calculates the minimum, maximum, and quartile values for the usage field.
Computing response time statistics by group
```
table duration=1h web_logs
| boxplot elapsed by method
```
Calculates the minimum, maximum, and quartile values of the elapsed field for each group sharing the same method field value.
Computing daily statistics
```
table duration=7d web_logs
| eval day = string(_time, "yyyy-MM-dd")
| boxplot bytes by day
```
Groups records by the day field and calculates box plot statistics for the bytes field.