All metrics

You can get a view of instance resources from the instance card.

You can also monitor the following metrics of an instance from the Metrics tab. Then from the tabs at the top, you can navigate between resources, instance and database metrics.

The metrics are laid out across three tabs according to their category:

  • Resources - Overall system resources, such as CPU, RAM and disk usage.

  • Instance - Information about the Neo4j instance(s) running the database.

  • Database - Metrics concerning the database itself, such as usage statistics and entity counts.

metrics

Select the info icon to see information about a particular metric, and you can expland the graph.

moreinfo

When viewing metrics, you can select from the following time intervals:

  • 6 hours

  • 24 hours

  • 3 days

  • 7 days

  • 30 days

Chart interactions

Memory and storage charts can be toggled between absolute and relative values using the % toggle.

Toggle data series

To hide or show individual data series, select the corresponding data series in the legend below the chart.

Zoom

To zoom in to a narrower time interval, select and drag inside any chart to select your desired time interval. The data will automatically update to match the increased resolution.

To reset zoom, double-click anywhere inside the chart or use the option in the context menu.

Expand

Any chart can be expanded to take up all the available screen estate by clicking the expand button (shown as two opposing arrows). To exit this mode, click the x button on the expanded view.

Context menu

To access the chart context menu, select the …​ button on any chart.

  • More info - Selecting More info brings up an explanation of the particular metric. For some metrics it also provides hints about possible actions to take if that metric falls outside the expected range.

  • Reset zoom - If the zoom level has been altered by selecting and dragging across a chart, Reset zoom resets the zoom back to the selected interval.

Aggregations

Most metrics will have several values for a given timestamp because of the following reasons:

  • Multiple database replicas

  • Compressing several data points into one, depending on zoom level

Aggregating functions are used to reconcile metrics having multiple data points and make the most sense of that particular metric. To convey an even more detailed picture of the state of the system, several aggregations can be shown.

The possible aggregations are:

  • Min - The minimum value of the metric across all cluster members.

  • Max - The maximum value of the metric across all cluster members.

  • Average - The average value of the metric across all cluster members.

  • Sum - The sum of the metric across all cluster members.

Detail view

AuraDB Business Critical AuraDB Virtual Dedicated Cloud

An Aura instance can run on multiple servers to achieve availability and workload scalability. These servers are deployed across different Cloud Provider availability zones in the user-selected region.

Detail view shows distinct data series for availability zone & instance mode combinations. This is presented as an alternative to the aggregations described above.

Detail view can be enabled with the toggle under the time interval selector.

Metrics in the Detail view for a new Aura instance may take time to appear because of the way 'availability zone' data is collected.

Store size metrics

Resources tab

The chart on the Resources tab shows the allocated store size metric for the selected database either as a percentage of the available storage assigned for the database or as absolute values.

Database tab

The Database tab provides a chart that shows the store size and the portion of the allocated space that the database is actively utilizing. Both metrics are represented as percentages of the available storage assigned to the database.

These metrics may differ due to the way Neo4j allocates and reuses space. Once allocated, space is never automatically de-allocated. Thus, reducing the data (nodes, relationships, properties) stored in the database does not reduce the top-line store size metric. However, Neo4j will reuse this 'available' space before allocating more from the system. The amount of allocated space that is 'available' is reported by the database, and Advanced metrics uses this metric to derive the used space by subtracting it from the allocated store size. This information can help you understand how close your database is to exceeding the assigned storage size.