| Paper | Title | Page |
|---|---|---|
| MOCOBAB05 | How to Successfully Renovate a Controls System? - Lessons Learned from the Renovation of the CERN Injectors’ Controls Software | 43 |
|
||
| Renovation of the control system of the CERN LHC injectors was initiated in 2007 in the scope of the Injector Controls Architecture (InCA) project. One of its main objectives was to homogenize the controls software across CERN accelerators and reuse as much as possible the existing modern sub-systems, such as the settings management used for the LHC. The project team created a platform that would permit coexistence and intercommunication between old and new components via a dedicated gateway, allowing a progressive replacement of the former. Dealing with a heterogeneous environment, with many diverse and interconnected modules, implemented using different technologies and programming languages, the team had to introduce all the modifications in the smoothest possible way, without causing machine downtime. After a brief description of the system architecture, the paper discusses the technical and non-technical sides of the renovation process such as validation and deployment methodology, operational applications and diagnostic tools characteristics and finally users’ involvement and human aspects, outlining good decisions, pitfalls and lessons learned over the last five years. | ||
|
Slides MOCOBAB05 [1.746 MB] | |
| MOPPC142 | Groovy as Domain-specific Language (DSL) in Software Interlock System | 443 |
|
||
| The SIS, in operation since over 7 years, is a mission-critical component of the CERN accelerator control system, covering areas from general machine protection to diagnostics. The growing number of instances and the size of the existing installations have increased both the complexity and maintenance cost of running the SIS infrastructure. Also the domain experts have considered the XML and Java mixture for configuration as difficult and suitable only for software engineers. To address these issues, new ways of configuring the system have been investigated aiming at simplifying the process by making it faster, more user-friendly and adapted for a wider audience. From all the existing DSL choices (fluent Java APIs, external/internal DSLs), the Groovy scripting language has been considered as being particularly well suited for writing a custom DSL due to its built-in language features: Java compatibility, native syntax constructs, command chain expressions, hierarchical structures with builders, closures or AST transformations. This paper explains best practices and lessons learned while building the accelerators domain-oriented DSL for the configuration of the interlock system. | ||
|
Poster MOPPC142 [0.510 MB] | |
| MOPPC143 | Plug-in Based Analysis Framework for LHC Post-Mortem Analysis | 446 |
|
||
|
Plug-in based software architectures are extensible, enforce modularity and allow several teams to work in parallel. But they have certain technical and organizational challenges, which we discuss in this paper. We gained our experience when developing the Post-Mortem Analysis (PMA) system, which is a mission-critical system for the Large Hadron Collider (LHC). We used a plugin-based architecture with a general-purpose analysis engine, for which physicists and equipment experts code plug-ins containing the analysis algorithms. We have over 45 analysis plug-ins developed by a dozen of domain experts. This paper focuses on the design challenges we faced in order to mitigate the risks of executing third-party code: assurance that even a badly written plug-in doesn't perturb the work of the overall application; plug-in execution control which allows to detect plug-in misbehavior and react; robust communication mechanism between plug-ins, diagnostics facilitation in case of plug-in failure; testing of the plug-ins before integration into the application, etc.
https://espace.cern.ch/be-dep/CO/DA/Services/Post-Mortem%20Analysis.aspx |
||
|
|
Poster MOPPC143 [3.128 MB] | |
| TUPPC028 | The CERN Accelerator Logging Service - 10 Years in Operation: A Look at the Past, Present, and Future | 612 |
|
||
| During the 10 years since it's first operational use, the scope and scale of the CERN Accelerator Logging Service (LS) has evolved significantly: from an LHC specific service expected to store 1TB / year; to a CERN-wide service spanning the complete accelerator complex (including related sub-systems and experiments) currently storing more than 50 TB / year on-line for some 1 million signals. Despite the massive increase over initial expectations the LS remains reliable, and highly usable - this can be attested to by the 5 million daily / average number of data extraction requests, from close to 1000 users. Although a highly successful service, demands on the LS are expected to increase significantly as CERN prepares LHC for running at top energy, which is likely to result in at least doubling current data volumes. Furthermore, focus is now shifting firmly towards a need to perform complex analysis on logged data, which in-turn presents new challenges. This paper reflects on 10 years as an operational service, in terms of how it has managed to scale to meet growing demands, what has worked well, and lessons learned. On-going developments, and future evolution will also be discussed. | ||
|
|
Poster TUPPC028 [3.130 MB] | |
| THPPC058 | LSA - the High Level Application Software of the LHC - and Its Performance During the First Three Years of Operation | 1201 |
|
||
| The LSA (LHC software architecture) project was started in 2001 with the aim of developing the high level core software for the control of the LHC accelerator. It has now been deployed widely across the CERN accelerator complex and has been largely successful in meeting its initial aims. The main functionality and architecture of the system is recalled and its use in the commissioning and exploitation of the LHC is elucidated. | ||
|
|
Poster THPPC058 [1.291 MB] | |