Experimental Physics and
Industrial Control System

"Elder Matias" <[email protected]> · Tue, 2 Sep 2008 18:05:31 -0600

At the Canadian Light Source we have been using RTEMS for quite some time.  It was a project Eric Norms started when he was with us in the late 1990s.  We currently use RTEMS in two configuration: (1) with EPICS running directly on RTEMS and (2) standalone transferring data to a second EPICS based machine.  In addition to the CLS the physics department runs RTEMS in an non-EPICS based system.

We have quite a few (aprox. 100) small (Motorola 68360 series) single board computers embedded in various pieces of equipment as well as used in IOCs for serial line devices.  These devices were first deployed in the 1999 time frame to around 2003/4.  Most  are still in use and we have done two EPICS upgrades over the years with only minor modifications.  

Orbit correction is also implemented on RTEMS.  In our configuration all the VME crates with IO associated with orbit correction are tied back to a single industrial (Intel based) PC using SIS optical VME-PCI bridges (SIS 3100).  We were a bit concerned about performance so chose to off-load EPICS onto another PC.  All the hard-real-time is done on the RTEMS machine and data is passed up to a second machine that then makes the data available over EPICS.  We were probably overly conservative and would have done equally as well with EPICS running on the orbit correction machine at a lower priority to the signal conditioning and control algorithms.

In both cases we have found RTEMS to be very reliable and stable.  There is a shell available, dynamic loading and remote debugging using the gnu tools.  We don't have any locally developed VxWorks system in use that makes it difficult for me to provide a head to head comparison.   

In addition to the CLS and completely independently of EPICS, the UofS physics department had ported the old in-house data acquisition system (called Lucid) that was used for nuclear physics experiments over to running on RTEMS a few years ago.  At the time the CAMAC hardware was replaced with VME and the same type of VME-PCI bridges we use was deployed.  The experiments with this detector and the Lucid data acquisition software are run down at JLabs or Duke.  Since the port was tied to an MSc thesis these was a bit of effort put into characterizing the real-time performance of the system.

One of the papers that came out of this project contains some benchmarks that may be of interest: http://ieeexplore.ieee.org/iel5/10384/33117/01557438.pdf

Daron Chabot use to be involved in the Lucid project and is now with us working on upgrading our RTEMS/EPICS based orbit correction and can provide a bit more technical details on RTEMS in both environments.

Elder

-----Original Message-----

To: "GOURNAY Jean-Francois" <[email protected]>, <[email protected]> 
Subject: RE: RTMES vs VxWorks 
From: "Dalesio, Leo" <[email protected]> 
Date: Tue, 2 Sep 2008 10:28:38 -0400 
References: <[email protected]> 
NSLS II at Brookhaven expects to use RTEMS. We are just starting - so you are further along than us. LCLS is using it and Till and Ernest should comment on that. Eric should say to what extent it is deployed at APS. And Elder should talk about their experience at CLS - I think that they have the longest history there - when Eric did the port of EPICS.
Bob

-----Original Message-----
From: [email protected] on behalf of GOURNAY  Jean-Francois
Sent: Tue 9/2/2008 7:22 AM
To: [email protected]
Subject: RTMES vs VxWorks

Hello All,

We are in the design phase of a new accelerator project which will be built in the frame of the so-called  "broader approach of the ITER project". The goal of this accelerator (Eveda) is the validation of the solutions for the future high power accelerator IFMIF (International Fusion Materials Irradiation Facility). 

The control system will be based (of course) on Epics with VME IOCs. The question is which RT OS to use? We are users of VxWorks for a very long time for several projects and we were always very happy with this system. It's expensive and we are no more privileged customers of VxWorks but it's for sure a great product. As we saw some interest for RTEMS in the community, we did complete tests with Epics including the port of drivers for our favourite modules. All this was satisfactory, so  we are really now at a cornerstone... and we need some feedback of facilities using RTEMS. What are the advantages or drawbacks based on your experience? Reliability? Off-the-shelf availability of drivers (for PMC modules for example)? availability of great features of VxWorks like the global symbol table or the interactive functions call? Etc ...

Many thanks,

Jean-François

J.F. Gournay

CEA Saclay

IRFU/SIS

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System