After a running-in period which was expected to last about a year, the new central computer, the CDC 7600, can now handle more work than all the other computers at CERN put together.
A sector now giving cause for concern is that of magnetic tapes. Difficulties arise because they are so numerous and differ in format. The large quantity poses problems in machine operation, because the mass of information they transmit can saturate the information transfer channels. The diversity of format creates software problems, since each type has to be processed in a different way.
However, a solution seems to be emerging. Magnetic discs will help to deal with the tape bottleneck. At the end of October a third large capacity disc was connected to the 7600. It will be used to store the data from several tapes in frequent use and will considerably lighten the operators’ task. Furthermore, the load on the channel linking the tape units and central memory is considerably reduced.
• Compiled from texts on p333.
Meeting on Storage Rings
In Spring 1974 the first experiments are scheduled to start at DORIS (DOppel-Rlng-Speicher), the electron–positron double storage ring at DESY. While the last preparations for the 3 GeV storage rings were still in progress, some of its initiators discussed further possibilities for developing the machine’s physics potential. They concluded that the amount of money needed for a normal high-energy physics experiment would be sufficient to add the ability to carry out electron–proton collisions within the rings.
The original plans were to have a 1 GeV proton booster but it has now been decided to do the main acceleration of the protons in the DESY electron synchrotron. Protons will be injected into the synchrotron at 4 MeV by a Van de Graaff accelerator. In one second about 1011 protons will be accelerated to an energy of 2 to 4.5 GeV and stacked in the storage ring. It should be possible to accumulate up to 3 × 1013 protons. At the two intersecting regions they will collide with 5 × 1012 electrons circulating in the second storage ring, resulting in luminosities as high as 1031 per cm2 per s.
This will be an excellent opportunity to study machine physics problems which will be important for the next generation of e–p storage rings. Given the importance of these topics, from 8 to 12 October experts were invited to DESY from all of the laboratories in Europe and the United States where storage rings are being planned, are under construction, or are already operating.
The meeting confirmed that multi-laboratory discussions on the different machine problems are a necessary and stimulating exercise to exchange information and to check the practicability of future projects. The local stimulation at DESY may have been responsible, a few days after the meeting, for bringing the first electrons and positrons from the DESY synchrotron to the injection points of DORIS.
• Compiled from texts on pp314–342.
Each nine-track magtape used for data storage in the 1970s could hold about 100 MB, so CERN’s 40,000-tape headache measured around 4 on the TB (1012 bytes) scale. Today’s LHC experiments produce more than 3 GB of data per second, creating that headache every 20 minutes. Together, they generate more than 30 PB (30 × 1015 bytes) per year, which are distributed to 170 computing centres in 42 countries connected by the Worldwide LHC Computing Grid. The centres are ranked in tiers, from 0 (largest) to 4 (smallest). The two Tier-0 centres, the CERN Data Centre in Meyrin and the Wigner Data Centre in Budapest, are responsible for the long-term safekeeping of the first copy of the raw data. While all centres store a fraction of the data on disc for fast access during analysis, the archiving medium is still magtape. It is inexpensive and durable, and one of today’s compact cartridges can hold up to 5 TB, exceeding CERN’s 1970s data headache.