UPGRADING THE SPS OPERATIONAL SOFTWARE FOR THE LHC ERA THE SPS-2001 SOFTWARE PROJECT

In December 1996, it was decided to set up a project to re-engineer the CERN SPS Accelerator application software to improve the operational efficiency and to cover the specific needs of the SPS as LHC injector. The mandate of this project is to provide a first working version of the software in June 2001 and a final version in July 2003. It will make use of the emerging CERN accelerator control infrastructure amongst which the new Java Application Programming Interface (this conference). This presentation reviews the reasons why the project was launched, its scope and objectives, the project planning and the methods used to run the analysis. The first results of the analysis phase and their implications for the application software, for the control infrastructure and for the accelerator equipment software services will be presented. 1 MANDATE OF THE SPS-2001 PROJECT The application software used today to control the SPS is the result of 20 years of evolution of the accelerator and computer technology. Many applications have been designed for and by equipment specialists. Although parts of the application software are much appreciated by the operators, today software does not allow the most efficient exploitation of the SPS machine. Since the SPS will probably run for another 20 years, it has been decided to set up a project to do the re-engineering of the SPS application software to allow more efficient operation and to cover the additional needs for the LHC era, that is the usage of the SPS as LHC injector [1]. The purpose of the project is to provide an application software suite to operate the SPS and its transfer lines in the LHC era that is allowing LHC filling, fixed target operation and machine developments (MD's). An important benefit of this project is to provide a homogenous software suite to the control room. The mandate of the SPS-2001 project consists of the definition of the strategy to operate SPS in 2001 onwards, the definition of the strategy to implement the new software and the production of all the software that will be used by the operation crew to operate SPS from 2001 onwards. In order to provide a homogenous solution to the control room, the SPS-2001 project will specify the interface to all equipment that must be accessed for the operation of SPS, define the standard services that this equipment should provide, define a standard Man Machine Interface and specify general purpose services and their interfaces needed to build the SPS-2001 software; e.g. Data Management Service, Alarm Service, etc. The SPS-2001 project will establish a communication path with the equipment groups and other entities that might be affected by the project. In addition, it will communicate SPS-2001 specific needs and requirements to other relevant projects such as the PS/SL Controls Convergence Project and its sub-projects (i.e. Java API, Middleware, Timing Convergence, etc.) The SPS-2001 project will provide the strategy to operate SPS from 2001 onwards, the strategy to implement the new software, the SPS-2001 certification document, guidelines to produce equipment services as well as application software, a first working version of the application suite in June 2001 and a complete final version in July 2003. 2 THE LHC REQUIREMENTS The LHC machine has specific requirements on the SPS operations that are driving the SPS-2001 project. These are fast and flexible cycle change, high machine availability and new operational concepts. The most obvious one is what is called the Fast and Flexible cycle change. Changing from one cycle or supercycle to another is possible today in the SPS but takes a non negligible amount of time. The future SPS software should provide some facilities that will allow the change of cycle in a much more flexible and rapid way. This will allow high machine availability and switching from LHC filling to fixed target physics will be done smoothly and rapidly. But this implies that equipment should not include assumptions about the super cycle structure and equipment must manage resident cycles and that new functionality has to be implemented in cycle dependant equipment (such as load, unload, activate...). Switching from one mode of operation to another requires new equipment surveillance to ensure that equipment is in the desired state for future mode changes. Subscription mechanisms would give the future software efficiency and flexibility. 3 RELATION WITH OTHER PROJECTS The SLI project [2] provides all LHC requirements that should be taken into account by the SPS-2001 project. International Conference on Accelerator and Large Experimental Physics Control Systems 51 International Conference Accel rator and L rge Experim ntal Physics Control Systems, 1999, Trieste, Italy