Experimental Physics and
Industrial Control System

"Garrett D. Rinehart" <[email protected]> · Wed, 17 Nov 1999 08:13:55 -0600 (CST)

We use sdds exclusively here at IPNS... Unix and NT versions (data sharing made 
possible by sddsendian). The max sample rate is 1Hz, I believe. Anything slower 
than that is fine though. We've got some large log sets that get saved once 
every five minutes and others that run every ten minutes. I've also done some 
short-term diagnostic logs of fewer channels up to 1Hz. It's always worked 
flawlessly for me.

BTW: There is a flag for sddsmonitor which allows it to run "triggered" instead 
of "periodic". I've never used it, but it may be useful to those not wanting to 
miss an important even and also not wanting to save loads of data at a high 
rate. If I'm not mistaken, you give it a channel to watch and it saves the whole 
log set when that trigger happens.

Garrett Rinehart
Intense Pulsed Neutron Source
Argonne National Laboratory
9700 S. Cass Ave
Argonne, IL  60439
(630)252-6561

> To: [email protected], [email protected]
> Subject: Re: Channel Archiving
> Cc: [email protected]
> 
> The channel archiver was not meant as areplacement for SDDS. The
> toolset available in SDDS is more inline with Matlab, IDL, or
> other analysis tools. The channel archiver is meant to save data
> over time at the rate that the process can produce it - or at some
> rate that the facility wishes to keep it. I was under the impression
> that the SDDS data logging facility was somewhat rate limited to
> 1 Hz and that it was impracticle for saving Giga bytes of data over
> long periods of time. Is this so? In any case, it would be a worthwhile
> addition to the viewers available for the channel archiver to be able
> to produce SDDS format data in addition to the spreadsheet format that
> they already produce. There are currently three data viewers that are
> available in Beta release - xarr (unix/Motif), a windows NT plot similar
> to xarr, and a cgi server that runs through a browser. The latest version
> of the channel archiver, written by Kay Kasemir, runs under either UNIX
> or Windows and was tested on a Pentium 450 MHz PC collecting near 10K
> signals per second. At 10K, there is some loss due to channel access flow
> control being automatically turned on when the archiver is busy writing
> the data to disk.
> 
> Could someone please clarify the data rates for SDDS logging?
> 
> 	Thanks,
> 	Bob

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System