Aggregation and HAVING - ChronusQL

Grouping creates independent state

Each unique combination of GROUP BY fields receives independent aggregate state. Add a window function when that state must be bounded by event time.

One result per driver and windowChronusQL

SELECT driver,
       AVG(speed) AS avg_speed,
       MAX(rpm) AS max_rpm,
       COUNT(*) AS samples
INTO output
FROM telemetry TIMESTAMP BY ts
GROUP BY driver, TumblingWindow(Duration(second, 10))

HAVING filters aggregate results

Before aggregation

WHERE

Controls which source events contribute to state.

WHERE speed > 0

After aggregation

HAVING

Controls which completed groups are emitted.

HAVING AVG(speed) >= 100

Filter complete groupsChronusQL

SELECT driver, AVG(speed) AS avg_speed
INTO output
FROM telemetry TIMESTAMP BY ts
WHERE speed > 0
GROUP BY driver
HAVING AVG(speed) > 200

Aggregate functions

Function	Result
`COUNT(*)`	Number of events in the group.
`AVG(field)`	Arithmetic mean of numeric values.
`MIN(field)` / `MAX(field)`	Smallest or largest value.
`STDDEV(field)`	Standard deviation.
`PERCENTILE(field, p)`	Percentile at numeric fraction `p`.
`MEDIAN(field)`	Median value.
`MAD(field)`	Median absolute deviation.
`REGRESSION_SLOPE(x, y)`	Linear regression slope.
`REGRESSION_INTERCEPT(x, y)`	Linear regression intercept.