View machine metrics

On the Machines page of the zeuz control panel (ZCP), you can view metrics for machines that match your filter criteria. See Filter machines for more information. The metrics show how your machines are performing based on a time period that you select.

The Active Machines section of the Machines page shows different types of metrics for your selected machines.

Image: Active Machines section showing performance data for one machine.

Select the following options to view different metrics for machines matching your filter criteria.

OptionDescription
DATEPICKERSelect one of the following time periods for which to display performance and usage information for your machines:

  • Live (default): Current performance data at one minute intervals, for the last 5 minutes.

  • Today: Today’s performance data, starting at 00:00H your time, in 5-minute intervals.

  • Yesterday: Yesterday’s performance data, in 5-minute intervals.

  • Last 7 days: Average hourly performance for the last seven days.

  • Last 30 days: Hourly performance for the last 30 days.

  • Last 90 days: Hourly performance for the last 90 days.

  • Custom: Selecting Custom allows you to enter a date range using a pop-up calendar datepicker. Performance data is displayed for the selected date range on hourly intervals.

If you select a time period other than Live, the performance charts do not auto-refresh.
CPU (AVG)Shows the CPU usage for each machine, ranging from 0-100%.

You can use the metrics_machinecpuusage API to fetch current real time data. For more information see metrics_machinecpuusage.
MEMORY (TOTAL)Shows the total server memory usage (RAM) for all machines, in mebibytes (MiB).

You can use the metrics_machinememoryusage API to fetch current real time data. For more information see metrics_machinememoryusage.
TRAFFIC IN (TOTAL)Shows the total inbound data transfer network traffic to the machines, in bytes per second (B/s).

You can use the metrics_machinenetworkreceived API to fetch current real time data. For more information see metrics_machinenetworkreceived.
TRAFFIC OUT (TOTAL)Shows the total outbound data transfer network traffic from the machines in bytes per second (B/s).

You can use the metrics_machinenetworksent API to fetch current real time data. For more information see metrics_machinenetworksent.

The Active Machines section shows performance data for each machine in a different color, as in the following example:

Image: Chart showing average CPU performance data for each machine for the previous day, with a different color for each machine. The color key for the machines is to the right of the chart.

Interpreting the charts

Peaks in usage metrics can indicate times of high demand and high CCU. Regular patterns of peak usage can help you determine when you need to allocate more resources to cope with increased demand. View data for the past 30 or 90 days can help you spot any patterns. If the trend shows that performance peaks rarely exceed 70% or lower, this might indicate that you can reduce the number of machines on your allocation or reduce their specification level. The aim is to get performance as close to 100% as possible during peak demand, without going over. You can use cloud machines to provide extra capacity at times when peak usage demands more than 100% of your bare-metal machine allocation.

Random peaks in machine metrics might indicate external triggers that caused a surge in demand (for example, media coverage about your game or new features might have spiked player interest).


2021-oct-05 Page added with editorial review.


Last edited on: October 21, 2021 (adcfeaae)