CERN Accelerating science

This website is no longer maintained. Its content may be obsolete. Please visit for current CERN information.

next up previous
Next: Christmas Shutdown Up: cnl225.html Previous: Letters from/to the Editor

A "Farewell" Message from the CN Division Leader

  David Williams CN

At the end of this year my appointment as CN Division Leader comes to an end, and it seems appropriate to contribute a short article to the Computer Newsletter on this occasion.

If I add the four years (1985-1988) which I spent as full-time deputy to Paolo Zanella, then the DD Division Leader, to the eight years as head of DD (in 1989) and then CN (1990-1996), I will have spent 12 years in "computing and networks" management at CERN. Speaking personally I have always been in favour of people at CERN changing jobs and responsibilities at a reasonable frequency, and that is especially true in fields such as computing where the rate of technological change is very high. And I am looking forward to being able to work, at least for some time, in a more technical job, so that I can update my, by now very rusty, technical skills.

This past decade has seen enormous changes in computing and network services at CERN. If I had the time I could have tried to make a survey of those changes, but that is work that needs to be done carefully in order to make it reasonably complete, and I do not have that time at the moment. In general, it seems to me that the period can be characterised:-

(1) by the arrival and successful processing of the enormous volume of data from LEP,
(2) by the relatively smooth transition from mainframes offering highly reliable batch and desktop services to a fully distributed computing environment for the whole of CERN and its user community, and
(3) by the first phase of the shift from a Fortran-based procedural paradigm to an object-based paradigm for HEP software, and especially the CERN Program Library.

All organisations, large and small, are having to adapt in order to cope with the new world created by the spread of computing and the Internet into all aspects of their activities. CERN and the whole HEP community will be no exception. We now need to really get to grips with the challenge of improving the overall efficiency of CERN via an optimal use of fully distributed computing. At the same time, together with the LHC experiments, we need to prepare a solid computing and networking foundation for those experiments. These twin challenges make this, in my view, a good moment for a change of leadership.

The CERN Directorate is in the process of setting some very sensible overall guidelines for the evolution of computing throughout the whole organisation, which I am sure you will learn more about during the first half of 1997, and they have also chosen Juergen May to replace me from January 1997. Many of you will already know him well. For those who do not, he has had a distinguished career as an applied physicist at CERN, including a long period in ALEPH. He recently spent five years as a Director at DESY, where he was responsible, among other activities, for central computing. Returning to CERN he was Technical Coordinator of ALEPH and LHC-B, before being asked to take up his new role. I would like to take this opportunity to wish him much success in his new and challenging tasks.

On this occasion I would like to thank all those people who have collaborated with me during my time at the head of DD and CN. Of course that includes all of the staff in those divisions, whether those with management functions or those who "simply" had technical responsibilities. I know how hard very many of you have been working, and I have been very proud to be responsible for many services which I know were and are often at the leading edge of what is feasible, given the enormous and fascinating demands posed by the informatics needs of the particle physics community and by the resources that can realistically be made available. Not to be forgotten are the technical staff and management of companies providing CERN with services of many different types in order to keep our informatics services running. We know that CERN is a demanding customer, and we rely heavily on you to keep things running smoothly. My thanks go also to our "users" in other divisions at CERN, and in the physics community at large. One of the valuable things that you have done, and I hope will continue to do, is to provide criticism of the services which we provide. Well-informed analysis from people who understand some of our constraints is especially valuable, but all statements of user requirements are useful, particularly if they really specify what you think you need rather than simply lobbying for a specific technical solution. I should also like to thank the staff and managements in many other computer centres and networks for their valuable and ongoing collaboration. And finally my thanks go to the many colleagues with whom I have worked in the CERN and Research Sector managements.

next up previous
Next: Christmas Shutdown Up: cnl225.html Previous: Letters from/to the Editor