The order operator in Axiom Processing Language (APL) allows you to sort the rows of a result set by one or more specified fields. You can use this operator to organize data for easier interpretation, prioritize specific values, or prepare data for subsequent analysis steps. The order operator is particularly useful when working with logs, telemetry data, or any dataset where ranking or sorting by values (such as time, status, or user ID) is necessary.

For users of other query languages

If you come from other query languages, this section explains how to adjust your existing queries to achieve the same results in APL.

In Splunk SPL, the equivalent operator to order is sort. SPL uses a similar syntax to APL but with some differences. In SPL, sort allows both ascending (asc) and descending (desc) sorting, while in APL, you achieve sorting using the asc() and desc() functions for fields.

```splunk Splunk example | sort - _time ```
['sample-http-logs']
| order by _time desc

In ANSI SQL, the equivalent of order is ORDER BY. SQL uses ASC for ascending and DESC for descending order. In APL, sorting works similarly, with the asc() and desc() functions added around field names to specify the order.

```sql SQL example SELECT * FROM logs ORDER BY _time DESC; ```
['sample-http-logs']
| order by _time desc

Usage

Syntax

| order by FieldName [asc | desc], FieldName [asc | desc]

Parameters

  • FieldName: The name of the field by which to sort.
  • asc: Sorts the field in ascending order.
  • desc: Sorts the field in descending order.

Returns

The order operator returns the input dataset, sorted according to the specified fields and order (ascending or descending). If multiple fields are specified, sorting is done based on the first field, then by the second if values in the first field are equal, and so on.

Use case examples

In this example, you sort HTTP logs by request duration in descending order to prioritize the longest requests.

Query

['sample-http-logs']
| order by req_duration_ms desc

Run in Playground

Output

_time req_duration_ms id status uri method geo.city geo.country
2024-10-17 10:10:01 1500 user12 200 /api/v1/get-orders GET Seattle US
2024-10-17 10:09:47 1350 user23 404 /api/v1/get-products GET New York US
2024-10-17 10:08:21 1200 user45 500 /api/v1/post-order POST London UK

This query sorts the logs by request duration, helping you identify which requests are taking the most time to complete.

In this example, you sort OpenTelemetry trace data by span duration in descending order, which helps you identify the longest-running spans across your services.

Query

['otel-demo-traces']
| order by duration desc

Run in Playground

Output

_time duration span_id trace_id service.name kind status_code
2024-10-17 10:10:01 15.3s span4567 trace123 frontend server 200
2024-10-17 10:09:47 12.4s span8910 trace789 checkoutservice client 200
2024-10-17 10:08:21 10.7s span1112 trace456 productcatalogservice server 500

This query helps you detect performance bottlenecks by sorting spans based on their duration.

In this example, you analyze security logs by sorting them by time to view the most recent logs.

Query

['sample-http-logs']
| order by _time desc

Run in Playground

Output

_time req_duration_ms id status uri method geo.city geo.country
2024-10-17 10:10:01 300 user34 200 /api/v1/login POST Berlin DE
2024-10-17 10:09:47 150 user78 401 /api/v1/get-profile GET Paris FR
2024-10-17 10:08:21 200 user56 500 /api/v1/update-profile PUT Madrid ES

This query sorts the security logs by time to display the most recent log entries first, helping you quickly review recent security events.

  • top: The top operator returns the top N records based on a specific sorting criteria, which is similar to order but only retrieves a fixed number of results.
  • summarize: The summarize operator groups data and often works in combination with order to rank summarized values.
  • extend: The extend operator can be used to create calculated fields, which can then be used as sorting criteria in the order operator.

Good afternoon

I'm here to help you with the docs.

I
AIBased on your context