Server metrics

GrapheneDB server metrics will allow you to track server errors and also request response times, as well as visualize incoming query and request throughput.



Server metrics are part of the Insights bundle, which is available for production-ready databases: Standard and higher, and Quartz and higher in Heroku

Data resolution and time windows

The measurements displayed in Metrics are aggregated in buckets of 15 minutes. To change the time window please click on the drop-down that is above the charts.

The available time windows are:

  • Last 3 hours
  • Last 6 hours
  • Last 12 hours
  • Last day
  • Last 3 days
  • Last week


Selecting a time window

Please note that Standard users time window is fixed to 3 hours.


Server errors are displayed in the chart as single red dots.

The size of the red dot will depend on the number of errors in the sampling interval window. Please read more about data resolution and sampling interval windows here.

If you hover over the error dot, additional information will be displayed.

Response time

Response time is displayed as a single stacked plot, combining the median (50th percentile) and the 95th percentile of the response time for the selected time window.

Median response time
The median response time displays the 50th percentile of HTTP requests (also Bolt) within the sampling interval, meaning that 50% of the requests to the database were completed within less time than the median, and 50% were completed within more.

95th response time
The 95th percentile response time of HTTP requests (also Bolt) within the specified sampling interval, meaning that 95% of the requests to the database were completed within less time than the median, and 5% were completed within more.


Number of requests served per minute for the selected time window.


Number of queries hitting the database per minute. Please note that a request can contain one or more queries.

Updated about a year ago

Server metrics

Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.