Experimental Physics and
Industrial Control System

Kay-Uwe Kasemir <[email protected]> · Fri, 21 Oct 2005 09:38:29 -0400

On Oct 20, 2005, at 18:01 , Andrew Johnson wrote:

* update on change in value
 with relative and absolute deadbands,
 each as a positive and negative limit.
 E.g. update when value increases by >= 1% or descreases by <=0.5%.
 EPICS V3 only has a single absolute deadband.

...

* update on archive related changes
 which are designated relative and absolute deadbands,
 again pos. & neg.,
 and in addition a minimum update rate.
 EPICS V3 only ADEL, no additional periodic update.

"Designated" by who? In V3 the ADEL and MDEL fields allows the DB

designer to set appropriate per-record deadbands; just allowing

every client to set its own deadband in the subscription request

doesn't provide any way for the DB designer to even hint what an

appropriate deadband might be, which could signficantly impact IOC

performance. Maybe that information doesn't belong in the database

at all, but I'm not convinced of that; I don't think we have all

the requirements for V4 deadbands specified yet.

What the requirements are is of course a fundamental question
and part of our current dilemma.

If we consider support for other systems like Tango a requirement,
then update rates and deadbands can be designated by each client.
Bob's "V4 Functional Specifications" presentation from the SLAC
EPICS meeting also included
"V4 Data Acquisition Capabilities: New subscription parameters:
rate limit, value changes (as before but also % change)".

So from that I'd conclude:
V4 and Tango support both require
deadbands for 'on change' updates designated by the subscribing client.

You're follow-up concerns are all valid,
but they are implementation details to be determined
once we know if per-client deadbands & rates are requirements.

As far as I understand, the outcome of the Archamps meeting is
that there's very little management support for V4 development
because of the impression that we've been designing pieces of
a neat implementation, but without sufficient agreement amongst
the developers and more important without requirements from customers.
So more than one 'plug' has been pulled and we're back to
use cases, then a 'client' API design from Cosylab, ...

-Kay

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System