OpManager MSP probe-to-central bandwidth consumption   

How is it calculated?

Bandwidth consumption is calculated based on the number of polls executed by a probe per minute. Please be informed that number of polls does not have a direct correlation with the number of devices you manage. It is affected by three major parameters such as
1) device type
2) frequency of polling and
3) enabled monitors.

Note: If you were the OpManager probe, you know that a desktop is different from an exchange or an active directory server, because the later are monitored for more parameters. Moreover, for you, a router is different from a server because a router is monitored more frequently than a server. (ex: a 5 min poll interval for router vs 30 min poll interval for a server). And finally your load varies with how many monitors are enabled for each device. (ex: server1 has only cpu monitor enabled where as server2 has cpu, memory, disc enabled)


How do i calculate number-of-polls-per-minute?

OpManager provides a UI for this. Go to central server ->admin->probes page and click on any probe name. You should be seeing a screen such as the one shown below. In this screen look for a field called probe load. In the screen below it shows 3.17 polls / minute and it monitors 24 desktops.

So whats the formula?

By experimenting several times we found that for 60 polls per minute the bandwidth utilization for a probe is approximately 2.83 MB per hour. You can exptrapolate this data for your probe load number. For example if the probe load is 3polls per minute then approximately 0.14 MB per hour.