HyperDX allows you to graph and aggregate any event you send to it. You can create dashboards to visualize your data and share them with your team.


Create a Dashboard

You can get started creating a dashboard by clicking "+ New Dashboard" link below the Dashboards section of the app navigation.

NOTE: By default, dashboards are not saved until you choose to explicitly save them. Unsaved dashboards can be shared via their URL to other teammates or for later use.

You'll be able to create multiple charts on a dashboard. Each chart can be resized and moved around as desired.

We recommend that important dashboards should be saved after creation by clicking the "Save Dashboard" button on the top right of the Dashboard page.

Create a Chart

You can create a chart by clicking the "Add Chart" button on the top right of any dashboard.

A chart can be created with the following options:

  • Name: The name of the chart, displayed within the dashboard.
  • Aggregation Function: The function to use to aggregate the data. For example Count of Lines, Sum, Average, 95th Percentile, Min, and Max.
  • Field: The field to aggregate on (when the aggregation function is not count).
  • Where: The filter condition to use to filter the data to be graphed. For example level:err or duration:>5000.
  • Group By: The field to group the data by. For example level or http.route

For example if you want to graph the average duration of error events grouped by span_name, you'd set the following options:

dashboard example

Examples of Common Charts

Here are a few examples of common charts you can create in HyperDX:

Count of Error Events Grouped by Service

  • Aggregation Function: Count of Lines
  • Where: level:err
  • Group By: service

95th Percentile Latency of Span Events Grouped by Span Name

  • Aggregation Function: 95th Percentile
  • Field: duration
  • Group By: span_name

HTTP Responses Grouped by Status Code

  • Aggregation Function: Count of Lines
  • Where: span.kind:server
  • Group By: http.status_code

http response grouped by status code example

Hi, how can I help you?