View project-specific packet loss dashboard

Stay organized with collections Save and categorize content based on your preferences.

Performance Dashboard packet loss metrics are automatically collected for your project.

You do not have to do any setup, but you do need a certain number of VMs in your project to get good results for the project performance view. For more information, see Metrics.

Events affecting Performance Dashboard packet loss metrics might take up to 10 minutes to appear.

To view packet loss project metrics, click the Packet loss tab on the Performance Dashboard page.

View project-specific packet loss information

The packet loss graph shows packet loss over time. The vertical axis shows the percentage of packet loss, and the horizontal axis shows when the losses occurred.


  1. In the Google Cloud console, go to the Packet loss page.

    Go to Packet loss

  2. From the Locations menu, select the regions for which you want to view data. You can select up to five regions. By default, if you do not select the regions, the information is shown for all regions.
  3. Choose a view interval by using the buttons across the top of the graph, from 1 hour to 6 weeks.
  4. On the horizontal axis, click a time to show a heatmap of average packet loss between zone pairs at that time. Slide the time selector to show packet loss at different times.
  5. In the heatmap, click a particular square to see a bidirectional graph of packet loss between those two zones.

In Shared VPC service projects, individual service projects might not be able to access all the information. For example, suppose one service project has VMs only in one zone and another service project has VMs in a different zone only. If the two sets of VMs are exchanging traffic, each service project can access packet loss details for the traffic heading toward their VMs.

Similarly, in peered Virtual Private Cloud (VPC) networks that sit in different projects, each project can access packet loss information only for the traffic heading toward their VMs.

To learn how you can view information from more than one project, see Network performance for multiple projects.

What's next