1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 <2018> 2019 2020 2021 2022 2023 2024 2025 | Index | 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 <2018> 2019 2020 2021 2022 2023 2024 2025 |
<== Date ==> | <== Thread ==> |
---|
Subject: | Re: EPICS for CompactRio |
From: | "Johannes Spinneken (EGI)" <[email protected]> |
To: | "Paduan Donadio, Marcio" <[email protected]>, "[email protected]" <[email protected]> |
Date: | Fri, 8 Jun 2018 18:02:39 +0000 |
One other observation on using the EPICS I/O servers on the cRIO. If communication speed is of concern, I would not recommend the NI I/O server. The I/O serves are not meant to provide real time capability. At the very best, you will be
able to achieve a 2-3ms data round trip efficiency, which is limited by the NI shared variable engine. In practice, you are more likely to see around 10ms data round trips between your cRIO and your other EPICS infrastructure. There is quite a lot of buffering
going on, particularly in the way in which the NI engine flushes its buffers to the network. For example, small packets under 8kB will only be flushed every 10ms, unless you force the flush via this VI: http://zone.ni.com/reference/en-XX/help/371361P-01/lvconcepts/sv_flush/ Best wishes Johannes Johannes Spinneken PhD FRSA Principal Systems Engineer Evergreen Innovations UK & USA +1 916 266 3709 (US cell) +44 7868 923583 (UK mobile) From: <[email protected]> on behalf of "Paduan Donadio, Marcio" <[email protected]> I would recommend you not to use NI shared variables architecture. I had a problem with arrays being read as a waveform record and discovered that NI implemented NORD and NELM concepts wrongly. In fact, one of this
fields does not even exist. I contacted the support and the answer was something like "we never mentioned 'IOC' or 'record' concepts in our material and, so, we do not expect that our solution will work exactly as an EPICS record". The solution
they gave to me was to make clear that the documentation will mention that they do not provide IOCs or records. Many of the fields you would expect in an EPICS record are not present. I suggest that, if you can, use the other solutions mentioned previously in this email thread. Best regards, Márcio Paduan Donadio Control Systems Engineer SLAC National Accelerator Laboratory Advanced Instrumentation for Research Division 2575 Sand Hill Rd Menlo Park CA 94025 Office (650) 926-5007 De: [email protected] <[email protected]> em nome de LOTRUS Paul <[email protected]> You have a good presentation on NI website:
http://download.ni.com/pub/branches/uk/handson_followup/epics_ral.pdf |