Hi,

I'm currently monitoring many (as in many hundreds) of devices, with an Advanced Python sensor collecting several channels of data from each device.

One of the channels reports the ambient temperature - I'm trying to figure out a way to report an average temperature reading from groups of devices (around 50 in each group).

I understand it would be possible to set up a Sensor Factory, adding each sensorID/channelNo manually, but that would be an extremely time consuming solution, since the number of devices monitored is going to continue to grow steadily into several thousand.

What I really need is a way to query the devices by group, tag, sensor type, etc. then calculate the average of a single channel value across all of the matching sensors.

Any ideas or assistance would be appreciated.

Article Comments

Hello,

I'm afraid to tell you that there's no native solution for this other than creating a factory sensor manually, sorry.


Jan, 2022 - Permalink

Thanks very much for the info, I’ve set one up, and it’s working ok.

I did notice the recommendation to keep the number of factory sensors below 50. Is this per-core, or per-probe? We would eventually need over 100 of these sensors once all of the devices are online, so it would be good to know if setting up extra probes will cover this.


Jan, 2022 - Permalink

All calculations from factory sensors are done by the core service. Since these calculations might have a high performance impact, we recommend to stay below 50 factory sensors per Core-Server. You can still try using more than 50 factory sensors and see how it goes however if you decide to do so, I recommend using a rather high scanning interval (300s and above) the reduce the overall performance impact.


Jan, 2022 - Permalink

Will do, thanks for the advice.


Jan, 2022 - Permalink