     |  |
This website is no longer maintained. Its content may be obsolete. Please visit http://home.cern/ for current CERN information.
|
ContentsEditorial If you need help Announcements Special 35th Anniversary Physics Computing
|
Computing at CERN, Once Upon a Time...Miguel Marquina (Editor) , IT/User Support AbstractA retrospective look at events in the CERN Computing arena. Five years ago...
At the present time, the UNIX Workstation Support team in CN/DCI Group offers support on a total of 7 UNIX architectures plus 2 flavours of X terminals. With ever-increasing demands on our services (the number of UNIX workstations has increased by some 50% over the past 2 years and the number of X terminals has at least doubled) and no prospect of increasing our resources, we are forced to look closely at limiting services... Our present efforts to rationalise platforms is not really new, is it? Thank Heaven Linux is there and there is light at the end of the tunnel. I mean RedHat Linux, not SUSE, or Caldera, or... I really mean RedHat 6.1. Or rather 7 by September? Surely Sun Solaris is a very acceptable exception to it. Sorry, I mean Solaris 2.6, not really 2.7 yet. But wait a minute, we cannot afford not plunging in parallel into it, can we? Rumours say Michael Ende might have found in this inspiration for his book... (Ed.) The structured cabling project is advancing well, with more than 5000 outlets installed and cabled, and more than 1800 users already connected. At the same time, the conversion to a routed network has made a good start, with the entire Computer Centre and several buildings converted so far. And today routed network is ubiquitous on-site. Another example of the foresight of the Computer Networks staff in offering us appropriate infrastructure to the information exchange needs of the lab for the years to come. A bow to Brian Carpenter from here my humble corner (Ed.) There are several typical mistakes that we are seeing regularly from users of the Listbox mailing list service. The first recommendation is to read the documentation! This and another four articles addressed topics on the then newborn Listbox service. It is interesting to see that the typical patterns we see in supporting users stand time just as solidly as rock. Just for the anecdote, did you know that I am the father of the service (built to replace the VM's LISTSERV), including its christening? Now you do. (Ed.) Ten years ago...
Up to now the public source files of the Program Library have been kept on binary PAM files... The Program Library Office is planning to make available and distribute only CARD PAM files in the standard areas. I managed to convince Julius Zoll The Legend to drop support for the "proprietary" binary format. And how proud I was... (Ed.) The KERNLIB package Z310 CFIO, to read/write UNIX files by interfacing to the C library was released [...] in March 1991. This worked correctly for disk files, but not for on-line tape files. Since SHIFT is now refreshed in our minds (see "Prestigious US Award for CERN Computing"), guess who created the ancestor of RFIO. Julius again! (Ed.) Twenty years ago...
Last autumn, the Scientific Policy Committee agreed that it was necessary for us to add a significant amount of capacity to the IBM system if we are going to be able to meet the increasing demands coming from the start of p-pbar experiments, increased intensity at SPS and from an expansion in the scale and range of terminal services. [...] The 3081 is a new processor from IBM that uses a new technology... It will have 16 Mbytes of memory 16 channels and two CPU's that together give a capacity of approximatively. 3.8 times that of the 370/168. [...] At the same time as the 3081 is installed the 3032 will be dismounted. This work will almost certainly take place over the weekend of Jeune Genevois, September 10-13. Two full mainframes relocated and practically operational in four days. CPU capacity blossoming to address computing requirements fully in the house. Those were different and challenging days. Today we trash a PC several factors faster and RAM-richer when a farm gets an itch. But we have other challenges equally stimulating. Find the ways to make LHC Computing a truly international Collaboration because we can no longer afford doing it alone. Experimental Physics keeps teaching us the ways... (Ed.) Thirty years ago and beyond...
The first edition (may it become historic) opened with the statement of intention: [see Macleod's quote in the Editorial] We hope the Newsletter in its first fifty editions has in some small way succeeded in this. Our attitude has been that the Newsletter should be just a little more than another boring trial, of the terminology that beggars any computer document. We regret that we have not found the time to present more information about the other computing activities of CERN. After all our department has fewer computers than either of the physics departments. Perhaps our regret is really that we have not succeeded in stirring up enough feeling about computers - for they dominate most people's lives and experiments at CERN. Letters have been few. We wish there had been more. By somebody who knew what Public Relations is about. (Ed.) |
|||||||||
