MQTT Dead Band

[ctm]

There is an old message similar to this dating to 2011 with a slightly different twist.  Let me explain.

In MQTT the concept of listener is important.  For me the dead band setting of any device therefore depends on the requirements of the listeners.

For example, if there are no listeners the dead band might be inifinity.  If there is only a chart recoder app listening the a dead band of 5 % will be adequate, etc.

To save storage and communication resources the MQTT server could manage this dead band by collating the requirements of the listeners and updating the dead band on the device accordingly.

Is there a mechanism in MQTT to achieve this?

[Nicholas O'Leary]

Hi,

I understand the general idea of a dead band, but could you explain what you mean by 'updating the dead band on the device'?

MQTT doesn't itself do anything to batch packets togther - which is what I think you are alluding to. As it runs over a tcp link, that sort of packet level management feels like it could be inserted beneath the MQTT layer...?

Nick

--
--
To learn more about MQTT please visit http://mqtt.org
 
To post to this group, send email to mq...@googlegroups.com
To unsubscribe from this group, send email to
mqtt+uns...@googlegroups.com
 
For more options, visit this group at
http://groups.google.com/group/mqtt
 
---
You received this message because you are subscribed to the Google Groups "MQ Telemetry Transport" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mqtt+uns...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[ctm]

I do not think there is a mechanism to achieve this in MQTT.

Maybe the best way to explain this is:  "If a tree in the forest falls but nobody hears, did it really fall?"

If nobody needs detailed information from a specific device, why should it report detailed information?   Devices reporting information not required, waste energy, bandwidth and storage space.

What I am suggesting is that MQTT finds a way of telling end point devices how often, and with what accuracy they should report.

C-: