[Halld-online] Interesting Communication from Cosylab

[Elliott Wolin]

Hi,

See especially their thoughts on using Labview with EPICS.

                                                                         
         Elliott



-------- Original Message --------
Subject: 	PAC follow-up
Date: 	Thu, 16 Jul 2009 16:07:49 +0200
From: 	Igor Verstovsek <igor.verstovsek at cosylab.com>
To: 	<wolin at jlab.org>
CC: 	'Klemen Zagar' <klemen.zagar at cosylab.com>


How are things at JLab? Do you have any budget for outsourcing that you could devote to collaboration with Cosylab? We have quite a lot to work on together! Below is a follow up of the discussion that we've had in Vancouver.

1) At PAC, you and Klemen were discussing the possibility of integration of LabVIEW into a control system (e.g., EPICS). One of the possibilities is to use the NI's proposed "Simple TCP Messaging" (STM) library [1], which allows communication with LabVIEW processes over an (open) TCP-based protocol

At Cosylab, we have developed a C and a Java library to communicate over STM, which you can find attached. With this, you can make any Java or C process (including EPICS IOC or client) talk with LabVIEW.

Note that this is presently a prototoype (e.g., we are aware of some race conditions when shutting down the C process which would need to be fixed for production use).

Note that there are also other alternatives now to interconnect EPICS and LabVIEW:

  * Sasha Zhukov's "native" implementation of EPICS CA in LabVIEW [2]
  * National Instruments' announced support for EPICS [3]

In any case, we believe that integration of LabVIEW makes a lot of sense for

  a) graphical user interfaces; these are really easy to construct with LabVIEW, and the results are of high quality. The GUI is then not as flexible as for custom-made GUIs (in C/C++, Python or Java), but if that
  is not a requirement, the time saving makes its use worthwhile

  b) real-time control loops (e.g., implemented with CompactRIO or LabVIEW real-time). Again, very easy to implement them, especially if the control loops are not very complex or require bleeding-edge performance.

2) You mentioned that presently you are improving your cMsg middleware to make it more redundant and fault-tolerant. At Cosylab, we have some experience with that (using group communication paradigm to implement active replication schemes; e.g., we participated at an EU-funded research project on the topic, [4]). Let me know if you would like some assistance, either in form of consulting or actual development work.

3) You might be interested to know that at Cosylab, we have decided to put some more effort into making EPICS more "industry-strength". What we mean by that is to package it in such a way that users would be able to download it and be up-and-running in a short time, without needing to hassle with looking for additional extensions and making them work together (in a similar paradigm we've made CAJ available, which allowed for hassle-free use).

We would be interested in your view on what extensions you see as a priority, and what you think EPICS distribution should be like. We know that an alarm system is high on your list, and we also plan to look in Kay's BEAST to see if we can package it in a reasonable way.  :-) 


				Sincerely,
					Elliott
 

================================================================================


 Those raised in a morally relative or neutral environment will hold
		    no truths to be self-evident.
				   

Elliott Wolin
Staff Physicist, Jefferson Lab
12000 Jefferson Ave
Suite 8 MS 12A1
Newport News, VA 23606
757-269-7365

================================================================================

-------------- next part --------------
A non-text attachment was scrubbed...
Name: LabVIEW-STM.zip
Type: application/octet-stream
Size: 410860 bytes
Desc: not available
Url : https://mailman.jlab.org/pipermail/halld-online/attachments/20090728/7c03fd73/attachment-0001.obj