Built for demanding analytical workloads.

Operational event analytics

The question: What is happening across active sessions now, where are flows failing, and which services need attention?
In RayoDB: Detailed events remain available for live service, session and reliability analysis as conditions change.
For: Operations, reliability and digital-service teams.

SELECT
  service,
  count(DISTINCT session_id) AS active_sessions,
  sum(event = 'error') AS failures
FROM events
WHERE ts >= '2026-09-04 10:00:00'
GROUP BY service
ORDER BY failures DESC;

Live activity and failures from detailed events.

Product and event analytics

The question: Which feature do users touch before they convert, and how does retention differ by channel, device and week?
In RayoDB: Joins and subqueries run over the raw events. Ask, look and refine without deciding the questions in advance.
For: Product and growth teams.

SELECT channel, toStartOfWeek(first_seen) AS cohort,
       count(DISTINCT user_id) AS active_week_4
FROM events e
JOIN first_touch f ON f.user_id = e.user_id
WHERE e.ts BETWEEN f.first_seen + INTERVAL 28 DAY
              AND f.first_seen + INTERVAL 35 DAY
GROUP BY channel, cohort;

Week-four retention by channel, without a cube.

Time series

The question: What will tomorrow look like, and which of these ten thousand series is misbehaving right now?
In RayoDB: Forecasting and anomaly detection are SQL functions over any series a query can produce.
For: Reliability and operations teams.

SELECT ts, value, score
FROM AI_DETECT_ANOMALIES(hourly_errors, hour, errors, 3.5)
WHERE is_anomaly;

SELECT * FROM AI_FORECAST(hourly_errors, hour, errors, 48, '0.1,0.5,0.9');

Anomaly scores and a forecast with its range.

Put your workload beside RayoDB.

Early access measures real queries, concurrency and infrastructure requirements.

Request access