Intelligence · Performance Insights
Trends that matter before they are incidents.
Saturation and growth analysis across database and system metrics, so capacity conversations happen on evidence rather than on the morning after.
What it analyses
Saturation, not just utilisation
A resource at 70% with a rising queue is a different situation from one at 90% and flat. Insights read the shape, not only the number.
Growth against the recoverable window
Data growth is read alongside backup duration and retention, because storage capacity and recovery capability run out together.
Per-instance baselines
What is normal for a reporting replica is not normal for a write primary. Trends are judged against the instance rather than a fleet-wide threshold.
Feeds the recommendation engine
A trend that crosses a threshold becomes a recommendation with its evidence attached rather than a chart nobody opened.
Related