InfluxDB in backend

Hello everyone,

I was wondering what’s the use of Timedata.InfluxDB in the backend? Is it the same as InfluxDB on the edge?
I’m asking because I activated Timedata.InfluxDB on the backend and connected it to the server, yet I don’t get any measurements.

Gratefully,
Ghada

Hello,
as far as i know, it is the same. but you need to connect a edge to the backend, so that data is logged.
best regards
Paul

Hello Paul,

thank you for your response, I wasn’t aware that I had to connect an edge to it!
So, to my understanding, the backend logs the data relative to the connected edge, and nothing more?

Thank you,
Ghada

Hei Ghana,

the backend is more or less a gateway for many edges and just for datalogging and display it. even if there is shortly no connection to the edge.
you can also connect via app over an backend to your devices.

As far as i understand it.

Actually i dont really use it.

what else do you expect that it logs?

best regards
Paul

Hi Ghada,

the OpenEMS Backend serves a couple of purposes right now:

  • Visualize and administrate an OpenEMS Edge via Cloud; this is useful when you do not have direct local ‘incoming’ access to an OpenEMS Edge. The Edge only needs outgoing access to the Backend on port 443, which is typically possible through every default firewall settings.
  • Storage of time-series data and visualization via UI.
  • Backend serves as a proxy relay between UI, Backend2Backend (B2B), etc. and Edge. As there are bidirectional websocket connections used, it is possible to communicate directly and instantly with the Edge. Backend provides a number of interfaces like B2B-Rest, B2B-Websocket and UI-Websocket that accect JSON-RPC Requests for that purpose.
  • It’s good to know, that for an OpenEMS UI it is completely transparent if it is connected to an Edge via Backend or to an Edge directly. This massively simplifies development of Cloud and Local Monitoring.
  • There are some developments going on on the backend side, like ‘Inbetriebnahmeassistent’, ‘App-Center’ and a ‘Notification Service’.

In a research project we also did some tests to distribute Virtual Power Plant (VPP) commands to different Edges, but that did not make it into the normal release. Right now the preferred way of implementing a VPP would be to do the calculation outside of OpenEMS Backend and set schedules e.g. via setGridConnSchedule (Backend-to-Backend communication :: Open Energy Management System). In fact this feature had been developed as per definition of a VPP provider.

Regards,
Stefan

2 Likes

hi @stefan.feilmeier , thank you for the clarification, it’s a bit clearer now.

Another question that I was wondering about, let’s say in a scenario where you have 2 edges deployed, each edge will be responsible for the balancing of the energy flow between the components connected to the edge in question, but how does the whole system (system composed of 2 edges, connected devices and the grid) balance itself? my hypothesis is that the balancing of the system is done via the controllers, but that is still a hypothesis as i didn’t find anything to support it.

Thank you,
Ghada