Experimental Physics and
Industrial Control System

[email protected] · Fri, 2 Sep 1994 15:09:24 -0700

> 
> I'm sorry but I forgot something...
> 
> 5. Assuming (let's say) M68040-class IOCs, what is the fastest loop that one
>    would normally close via database processing, ie via some sort of
> 
>    input record   --->   algorithm record   --->   output record
> 
>    chain? I just mean a ballpark figure. 1 Hz? 10 Hz? 20 Hz? Presumably one
>    needs to consider two cases:
> 
>    a) all above database records on same IOC
>    b) database records distributed across more than one IOC
> 
>    How much is the answer influenced by scan task priorities (assuming that
>    CPU is not a scarce resource)?
Hmm, Sounds like you need to read the SSC's EPICS Performance Study that was
done with M68040s.

I have a postscript copy and can email you a uuencoded version if you wish.

Aloha,
	Peregrine

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System