EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

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  <20212022  2023  2024  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  <20212022  2023  2024 
<== Date ==> <== Thread ==>

Subject: Re: Going full PVA?
From: Timo Korhonen via Tech-talk <tech-talk at aps.anl.gov>
To: Stainer Tom <Tom.Stainer at sckcen.be>, "tech-talk at aps.anl.gov" <tech-talk at aps.anl.gov>
Date: Tue, 8 Jun 2021 15:54:22 +0000

Hi Tom,

 

I guess it would be possible to create such a gateway, although I have not yet seen a case where that would be strictly necessary. I can imagine that there are such cases, we just have not seen any real blockers yet, or at least none have come to my knowledge.

Maybe other colleagues can give better comments to that issue. We try to be as pure PVA as possible in any case.

 

We do have some vendor devices that provide EPICS integration, but they provide other protocols as well that we can access with StreamDevice or other drivers.

 

I know that there are other sites working with PVA as well, I hope to hear comments from them as well.

 

But being where we are now, I would never want to go back to CA.

 

Best regards,

 

Timo

 

 

From: Stainer Tom <Tom.Stainer at sckcen.be>
Date: Tuesday 8 June 2021 at 15:53
To: Timo Korhonen <Timo.Korhonen at ess.eu>, "tech-talk at aps.anl.gov" <tech-talk at aps.anl.gov>
Subject: RE: Going full PVA?

 

Dear Timo,

 

Thank you for your detailed response, this is sort of what I was hoping for.

 

Indeed I had read somewhere (an old presentation I think) that at ESS the intention was to go full PVA, it just wasn’t clear if this was actually the case. Thanks for confirming.

 

What I meant with CA <-> PVA translators is if some device is providing data over CA is there a way to make a proxy or gateway to wrap this and convert it to a corresponding PVA signal? As you mentioned below, avoiding vendor specific EPICS integration is always desired but sometimes unavoidable and I was thinking in these cases what can you do if you want only PVA….. I wondered if by going “full PVA” you end up restricting yourself at some point due to legacy vendor devices, lack of support in modules/extensions, etc.

 

Knowing that ESS is aiming for full PVA is good to know. I appreciate the feedback.

 

Kind regards,

Tom

 

From: Timo Korhonen <Timo.Korhonen at ess.eu>
Sent: Tuesday, June 8, 2021 12:36 PM
To: Stainer Tom <Tom.Stainer at sckcen.be>; tech-talk at aps.anl.gov
Subject: Re: Going full PVA?

 

Hi Tom,

 

We at ESS are aiming to go full PVA. So far it is proceeding quite well, although it will take some time before we get to be 100% “CA-free” – if ever, even if that is still the goal.

I do not consider it too bad if we have some fringe uses of CA left (e.g., CA links in IOC databases) as long as our IOCs and the application level all runs on PVA. That we have pretty much achieved, there is some work left in updating legacy IOCs that have been developed before EPICS 7 was available. In theory one could even run a PVA server on those but we rather update them to more recent releases, for many other reasons besides PVA compatibility.

 

If your legacy device is an IOC (i.e., you are talking to that device via an IOC), then there are little or no issues. Just take a recent EPICS base and run your IOC on it. This obviously assumes that you have all the pieces available to recompile the IOC. But if you do, this is straightforward to do.

However, if you have an embedded IOC that is stuck to some older EPICS version, then it will be difficult. For this reason we avoid using vendor EPICS integrations that used to be popular some time ago and rather use StreamDevice/Modbus/asyn or similar to integrate those, and do EPICS integration on our side.

 

I am not aware of any CA<->PVA translators; we have not seen the need in our applications. I am not even sure how much sense that would make (IMHO).

 

As a new facility we have been able to start more or less from scratch, very little, if any, legacy to support, so we probably have had a simpler journey than most others would have.

 

Best regards,

 

Timo

 

From: Tech-talk <tech-talk-bounces at aps.anl.gov> on behalf of Stainer Tom via Tech-talk <tech-talk at aps.anl.gov>
Reply-To: Stainer Tom <
Tom.Stainer at sckcen.be>
Date: Tuesday 8 June 2021 at 09:52
To: "
tech-talk at aps.anl.gov" <tech-talk at aps.anl.gov>
Subject: Going full PVA?

 

Dear tech-talk,

 

I am looking for some general advice and expertise on CA vs PVA in practice. After reading about the differences and the theoretical points of view, I am now interested to learn about the practical side of things.

Specifically, if one were to go full PVA (no CA), what pitfalls or problems would I encounter?

 

For example, if I have to support a legacy system or device which already has a CA interface how difficult is it to port or map this to PVA?

Do CA <-> PVA translators exist in the community?

 

I am of course aware of certain extensions and libraries not yet supporting PVA, but my question relates more to the hardware side of things and networking.

 

Maybe an additional question which might prove helpful – has anyone ever gone full PVA?

 

The prospect of using one protocol is very much more appealing and PVA seems to be the new and improved protocol. Particularly I would like to take advantage of things like RPC and the “smart database”.

 

I realize open questions like these can sometimes yield some varied responses but any advice would be greatly welcomed.

 

Kind regards,

Tom


Replies:
Re: Going full PVA? Ralph Lange via Tech-talk
References:
Going full PVA? Stainer Tom via Tech-talk
Re: Going full PVA? Timo Korhonen via Tech-talk
RE: Going full PVA? Stainer Tom via Tech-talk

Navigate by Date:
Prev: RE: Going full PVA? Stainer Tom via Tech-talk
Next: cbLow consumption randomly stopping Daykin, Evan via Tech-talk
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  <20212022  2023  2024 
Navigate by Thread:
Prev: RE: Going full PVA? Stainer Tom via Tech-talk
Next: Re: Going full PVA? Ralph Lange via Tech-talk
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  <20212022  2023  2024 
ANJ, 09 Jun 2021 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·