Re: Re: History of EPICS

["Nonn, Patrick via Tech-talk" <tech-talk at aps.anl.gov>]

Hi,

thanks for all the replies. I now have plenty for my slides. Here is a little summary, for future reference:

1987: Los Alamos National Laboratory helped Argonne National Lab with the development of the control system of the Intense Pulsed Neutron Source, specifically by providing operator interfaces. This marked the first time, two national laboratories collaborated on the development of a distributed control system, instead of everybody just sticking to their home-brew.

1988: At the "Accelerator Application Automation Toolkit Workshop" control system experts from around the world (including Argonne National Lab) discussed the dos and don'ts of future control systems for accelerators. The outcome of those discussions were the foundation for the development of the "Ground Test Accelerator Control System (GTACS)" at Los Alamos. (Sidenote: The Ground Test Accelerator (GTA) was a test bed for developing accelerator technology to be used in the Strategic Defence Initative (SDI))

1989: Argonne required a control system for their Advanced Photon Source (APS). At the ICALEPCS'89 they met with the team from Los Alamos. They agreed to cooperate and Argonne send Marty Kraimer to Los Alamos for 6 months as part of that cooperation. When Marty went back to Argonne, he convinced his team, that GTACS could be used for the APS without the need to fork from it and develop their separate control system (which was usual at the time).

1991: The product of the collaboration between Los Alamos and Argonne was presented at the ICALEPCS'91. As the resulting control system was no longer specific for the GTA, it was rechristened "Experimental Physics and Industrial Control System (EPICS)". It started with the version (or release) number R3.1. The "major" version number reflects the version of the Channel Access protocol. Versions 1 and 2 were already deprecated, as EPICS came around. The "minor" version reflects the development of the IOC-core.

1991 to 1996: Various other accelerator labs around the world joined the collaboration (in no particular order: LBL, CEBAF(JLab), IPNS, DESY, SLAC, KEK). Also some observatories joined (KECK, Gemini). Three companies (Kinetic Systems, Tate Integrated Systems, Titan Corp.) bought the commercial license (back in the day Open Software was a new concept, bar any legal framework). In that time frame the versions "burned" through numbers R3.1 to R3.12. I can only speculate, that any contribution spawned its own IOC version, which was later combined in R3.11. Version R3.12 is the oldest version (or at least oldest tag), still in the repository.

1996: Version R3.13 (based on my review of the repository log) introduces support for non-VxWorks operating systems.

1996 to 2004: More Labs used EPICS (i.e.: SLS, SNS, DIAMOND, S-DALINAC, PSI, BESSY). Thus also the ecosystem of modules, extensions and supported devices grows. EPICS becomes the quasi standard for accelerator control systems.

2004: Thanks to Andrew Johnson, the licensing switches from a commercial license to a more open one. While not strictly Open Source, it is no longer required to register as a collaborator with Argonne, in order to use EPICS base free of charge.

2005: In a meeting at ANL the future of EPICS is discussed. Based on the ideas, coming up at that meeting, Marty Kraimer starts to develop the concept of an EPICS, able to communicate and process structured data.

2006: Marty leaves ANL and starts developing, what would become to be called EPICS 4. This development was done in Java.

2007: Matej Sekoranja (COSYLAB) joins Marty in the development of EPICS 4.

2010: The development of EPICS 4 switches to C++, in order to become more compatible with the (ongoing) EPICS 3 development. This attracts also more co-developers.

2017: The development of EPICS 3 continued parallel to that of EPICS 4, resulting in EPICS R3.16. The IOC core of EPICS 3.16 was combined with EPICS 4, resulting in EPICS 7. EPICS 4 was never released as a stand-alone. The development of R3.16 was closed.

Current: Two versions of EPICS are supported: 3.15.x and 7.0.x.

I would like to share some information from my neck of the woods:
We, at DESY MSK group, support EPICS for MicroTCA-based LLRF control systems at Helmholtz-Zentrum Berlin (HZB), Turkish Accelerator and Radiation Lab in Ankara (TARLA) and at Karlsruhe Institut für Technologie (KIT).

We have two different ways for an IOC to connect to a MicroTCA AMC via PCIe:

1. The EPICS Device Support: https://github.com/ChimeraTK/ControlSystemAdapter-EPICS-Device-Support.git

As the name suggests, this builds you a device support, which can than be added to your IOC.
2. ChimeraTK Application with IOC  adapter: https://github.com/ChimeraTK/ControlSystemAdapter-EPICS-IOC-Adapter.git

This would be linked with a ChimeraTK application, to generate a full fledged IOC.

Again, thanks for all the info.

Kind Regards,

Patrick Nonn

---

From: "Steven Hunt" <Hunt at lbl.gov>
To: "Maren Purves" <m.purves at eaobservatory.org>
Cc: "Mathew Rippa" <mrippa at gemini.edu>, "tech-talk" <tech-talk at aps.anl.gov>, "Patrick Nonn" <patrick.nonn at desy.de>
Sent: Wednesday, 26 February, 2025 04:16:38
Subject: Re: [EXTERNAL] History of EPICS

Ha .....

https://www.osti.gov/etdeweb/biblio/6579587

Also Just remembered we went to ICALEPS in Japan 1991, but can't remember if we were already using EPICS then.

On Tue, Feb 25, 2025 at 6:59 PM Steven Hunt <Hunt at lbl.gov> wrote:

I remember some very early meetings with Bob, Marty, and Jeff when I was at the SSC (Bubbatron) , I think held at LANL (AT-8).  This was  probably in 92/93 if someone could confirm? I don't know if Dave Gurd, or Doug Murray (now at SLAC) still read Tech-Talk, but I will try to reach out to them to see if they remember.  What I seem to remember is  we were advised, probably by Marty,  to use the 'standard' config using Sun workstations for clients (we used about every flavor of workstation, including Sun, DEC, HP and IBM-AIX). But I am pretty sure CEBAF at that time used HP.  Would Matt Stettler have been involved then? Would Joan Sage (formally CEBAF) also have been at SSC already then? Of course after this long, the timeline is hard to remember.

I found this in my notes, don't know how accurate it is, certainly not complete (GTA, SSC,...) the numeric values may have been the number of IOCs, this may have been only machines that were still running when I compiled this (probably when I was at ITER).   Might be interesting to complete this list, perhaps to add names of some of the people involved.

*ITER ITER EPICS Linux TCA? 500? 2017?
MedAustron EPICS? Linux? ? 50? 2013?
*Maxlab MaxIV EPICS? Linux? ? 100? 2013?
*BNL NSLSII EPICS Linux Home ? 2011?
PSI XFEL EPICS VxWorks VME 100 2010?
SLAC LCLS EPICS RTEMS VME 100 2008
*AusSyn AS EPICS Linux PCI 30 2007
SSRF SSRF EPICS Linux ? ? 2007
* RAL DLS EPICS VxWorks VME 300 2007
SSRL Spear3 EPICS RTEMS VME 100 2004
ORNL SNS EPICS VxWorks VME 300 2005
* PSI SLS EPICS VxWorks VME 250 2002
* CLS CLS EPICS Linux VME 100 2002
SSLS SSLS EPICS ? 2002
NSRC NSRC EPICS? 2001
BESSY BESSY2 EPICS VxWorks VME 200 2001?
KEK PF EPICS 1997
ANL APS EPICS VxWorks VME 300 199?
JASRI Spring8 EPICS 1996
NSRRC NSRRC EPICS 1993
* LBL ALS EPICS VxWorks VME 100 1991

On Tue, Feb 25, 2025 at 4:19 PM Maren Purves via Tech-talk <tech-talk at aps.anl.gov> wrote:

Just a minor addition:
I know that Mary Fuka came to JCMT from LANL but I do not know what
she worked on at LANL.
I was told (don't remember by whom of the two) that she and Peregrine
McGehee (now at SLAC, we worked at the Caltech Submm Observatory at
the same time, about 30 years ago) once worked on the opposite ends of
the same beamline :)

Maren

On Tue, Feb 25, 2025 at 10:47 AM Mathew Rippa <mrippa at gemini.edu> wrote:
>
> Hi Patrik,
>
> As Maren points out, UKIRT and Keck may have been the first Telescopes using EPICS 3.12-ish(?) in the 90's. I recall William Lupton (Keck) and Nick Rees(UKIRT,JCMT) spearheaded the initiative to bring it to the Astronomy community. This was later supported by Maren, Allan Honey (drvSerial/drvAscii) collaborating with Jeff Hill and later Kevin Tsubota (Keck). When the Gemini project started many of the RGO staff developed EPICS related work packages: VMIC-5588 Reflective Memory, PMAC1 support, AllenBradley PLC5, GenSub Record (Andy Foster, Andrew Johnson et al), Xycom 240/566 (now acquired by Acromag?). I believe Mary Fuka (JCMT) worked on the predecessor to EDD/DM called GDCT(spelling?). Does that sound right, Maren?
>
> In 2009, I looked at Eric Norum's work bringing RTEMS 4.x to EPICS on PPC targets. At the time Gemini had its standard control as the mvme167 (68k) and the mvme2700 (MPC75). With Till Strauman (SLAC at the time) and Eric's help, I got the 2700 building for RTEMS 4.9.2 and it was accepted in EPICS 3.13 circa 2009. Gemini has since phased out the 167's and the 2700 is the ubiquitous Gemini controller. The 2700 now supports RTEMS 6.1 and development branches EPICS 7.
>
> Aloha,
> -Matt
>
>
>
> On Sun, Feb 23, 2025 at 1:13 AM Foster, Andrew (Observatory Science, RAL, TEC) via Tech-talk <tech-talk at aps.anl.gov> wrote:
>>
>> Aloha Maren,
>>
>> No, Observatory Sciences never bought a commercial license. But we've been involved with EPICS since the start of the company in 1998. Some of us, before that, working at the Royal Greenwich Observatory, on Gemini, and various Canary Island telescope projects.
>>
>> Cheers, Andy
>>
>>
>> Sent from Outlook for Android
>> ________________________________
>> From: Tech-talk <tech-talk-bounces at aps.anl.gov> on behalf of Maren Purves via Tech-talk <tech-talk at aps.anl.gov>
>> Sent: Sunday, February 23, 2025 10:30:53 AM
>> To: Hartman, Steven <hartmansm at ornl.gov>
>> Cc: Nonn, Patrick <patrick.nonn at desy.de>; tech-talk <tech-talk at aps.anl.gov>
>> Subject: Re: [EXTERNAL] History of EPICS
>>
>> That was a great read - finally getting around to it now - , as was
>> Bob's later email reply, still missing Marty, and Steve Lewis as well.
>>
>> EDD/DM was written by Deb Kerstiens and I ported it to Linux after she
>> retired, when we (at the JAC, mostly working at UKIRT at the time,
>> about 2002) switched from Solaris to Linux. JCMT got EPICS later and
>> is using MEDM. UKIRT, now operated by the University of Hawaii
>> Institute for Astronomy, is probably still using EDD/DM unless their
>> part time software person has converted it all to Python interfaces.
>> We, now East Asian Observatory and only operating the JCMT, still run
>> EPICS 3.13.8 with some 3.14.7 (for camonitor) - and as time allows
>> taking some stabs at EPICS 7.
>>
>> Another one of the Maunakea Observatories using EPICS, these days on
>> RTEMS, is Gemini (at least North).
>>
>> As somebody asked about commercial licenses, did Observatory Sciences
>> buy a commercial license or was that after it went open source? Also,
>> Cosylab?
>>
>> Aloha,
>> Maren Purves
>> Head of Instrument and Telescope Software
>> East Asian Observatory / JCMT
>>
>> On Fri, Feb 21, 2025 at 11:04 AM Hartman, Steven via Tech-talk
>> <tech-talk at aps.anl.gov> wrote:
>> >
>> >
>> > On Feb 21, 2025, at 1:53 PM, Johnson, Andrew N. via Tech-talk <tech-talk at aps.anl.gov> wrote:
>> >  I might be able find an online history that the late Marty Kraimer started writing,
>> >
>> >
>> >
>> > Still available on GitHub . . .
>> >
>> > https://urldefense.us/v3/__https://eur03.safelinks.protection.outlook.com/?url="">
>> >
>> >
>> > --
>> > Steven Hartman
>> > hartmansm at ornl.gov
>> >
>> >
>> >
>> This e-mail and any attachments may contain confidential, copyright and or privileged material, and are for the use of the intended addressee only. If you are not the intended addressee or an authorised recipient of the addressee please notify us of receipt by returning the e-mail and do not use, copy, retain, distribute or disclose the information in or attached to the e-mail. Any opinions expressed within this e-mail are those of the individual and not necessarily of Diamond Light Source Ltd.
>> Diamond Light Source Ltd. cannot guarantee that this e-mail or any attachments are free from viruses and we cannot accept liability for any damage which you may sustain as a result of software viruses which may be transmitted in or with the message.
>> Diamond Light Source Limited (company no. 4375679). Registered in England and Wales with its registered office at Diamond House, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom.