Page tree
Skip to end of metadata
Go to start of metadata

Authors

P. Golonka, F. Varela Rodriguez

Oral or poster

Oral

TrackControl System Upgrades
Status

APPROVED ID 1685 GL APPROVED READY FOR DL WEDPL04

Innovation

Modernization of the frameworks; new functionality, concepts, areas of applicability; upcoming open source licensing effectively enables use beyond CERN
ApplicationIn addition to the CERN experiments accelerator sector and technical infrastructure, also GSI, ITER and potentially many other research insitutes and their industrial integrators, as interest grows in the use of UNICOS beyond CERN to implement industrial control systems or appliances
Complexity
  • Evolution of a software framework used as foundation for a large number of long-term projects put contradicting requirements of quality (stability), innovation and backward-compatibility at the same time. This calls for systematic development methodology and extensive testing. 

  • Large scale software project comprising almost 2 millions of lines of code written in various languages and covering a stack of complex technologies, developed over the last 18 years with a very high turnover in the team 

  • Very significant technical debt accumulated in the code due to the lack of proper previous QA measures, contributions from lots of temporary personnel, knowledge silos and knowledge loss in some critical areas. 

  • Transition to agile development with a team of few developers with high turnover – typical solutions such as SCRUM may not be directly applied.

Abstract

The Industrial Controls Frameworks, JCOP and UNICOS, have been employed to develop hundreds of critical controls applications in multiple domains like the LHC Experiments Detector Control System, accelerator complex (cryogenics, powering, interlocks) or technical infrastructure, leading to an unprecedented level of homogeneity. These frameworks, used by a thousand of developers worldwide, will now undergo a major consolidation and re-engineering effort to prepare them for the new challenges of the next 20 years in the HL-LHC era, as well as to streamline their maintenance. 

The paper presents the challenges that will be faced during this project due to the breadth of technological stack and large code-base contributed over two decades by numerous authors. Delivery of innovation induced by evolution of technologies and refactoring of the ageing code must be done in a way that ensures backward-compatibility for existing systems. The vision and the current state of the frameworks is discussed, alongside the main deliveries planned in the medium term. Lessons learnt, optimizations of processes to make best use of available resources and efforts towards open-source licensing of the frameworks are also presented. 


Paper

The paper was uploaded to EDMS: https://edms.cern.ch/document/2231097

Below is a local copy for reference

2 Comments

  1. Abstract may reflect instead of 650 critical applications the kind of control systems (e.g. cryogenics, detector control systems) and emphasises some numbers (e.g. experiments manpower developing on top of the framework...)

    challenge: living control systems for the LHC life...

    reduced team to develop the fw? → methods

    collaborative approach... → open source (community)



  2. Reviewed, I have some minor comments in a printed copy