Language

Aggregation

ChronusQL supports finite batch aggregations and event-time windowed aggregations.

Guide

ChronusQL supports finite batch aggregations and event-time windowed aggregations.

Global Aggregation

sql
SELECT AVG(server.memory) AS AvgMemory INTO output
FROM input

Tested input:

jsonl
{"server":{"memory":20}}
{"server":{"memory":40}}
{"server":{"memory":10}}

Tested output value:

json
{"AvgMemory":23.333333333333332}

Grouped Aggregation

sql
SELECT server.region AS Region, AVG(server.memory) AS AvgMemory INTO output
FROM input
GROUP BY server.region

Tested behavior:

  • `us-east` memory values `20` and `40` aggregate to `30`.
  • `us-west` memory value `10` aggregates to `10`.

Multiple group keys are supported:

sql
SELECT server.region AS Region, server.cluster AS Cluster, COUNT(*) AS EventCount INTO output
FROM input
GROUP BY server.region, server.cluster

Aggregate Functions

The aggregation tests cover:

  • `COUNT(*)`
  • `AVG(field)`
  • `MIN(field)`
  • `MAX(field)`

Example:

sql
SELECT server.region AS Region, MIN(server.memory) AS MinMemory, MAX(server.memory) AS MaxMemory INTO output
FROM input
GROUP BY server.region

Nested output aliases are supported:

sql
SELECT AVG(server.memory) AS [metrics.avg] INTO output
FROM input

Many Aggregate Expressions

The tests validate fifteen aggregate expressions in one query:

sql
SELECT
    AVG(v1) AS avg_v1, MIN(v1) AS min_v1, MAX(v1) AS max_v1,
    AVG(v2) AS avg_v2, MIN(v2) AS min_v2, MAX(v2) AS max_v2,
    AVG(v3) AS avg_v3, MIN(v3) AS min_v3, MAX(v3) AS max_v3,
    AVG(v4) AS avg_v4, MIN(v4) AS min_v4, MAX(v4) AS max_v4,
    AVG(v5) AS avg_v5, MIN(v5) AS min_v5, MAX(v5) AS max_v5
INTO output
FROM input
PreviousTimestamp HandlingNextWindowing