China 
Africa 
Explore chapters and articles related to this topic
Computer Technology Primer
Published in Michael M. A. Mirabito, Barbara L. Morgenstern, Mitchell Kapor, The New Communications Technologies, 2004
Michael M. A. Mirabito, Barbara L. Morgenstern, Mitchell Kapor
Application programs are created through programming languages. A programming language is considered a control language. It provides the computer with a set of instructions to perform a series of operations.
Theoretical spectroscopy in the early days of digital computing – an homage to Sigrid D. Peyerimhoff
Published in Molecular Physics, 2020
The perfomance of the programme must, of course, be seen in the historical context of the very limited digital computer resources in the 1970ies. To save random access memory (RAM), computer codes with complicated overlay structures had to be devised. Software limitations – dynamical allocation of memory for arrays and matrices was not possible in FORTRAN then and required calls to assembler routines – forced the programmer to reuse the workarray space over and over again which made the codes very difficult to grasp. From a modern point of view, the programming style was horrible, but at that time this type of memory management was inevitable. The codes were punched on cards and each user had to carry a card deck containing the programme and the input as well as complicated job control language to the computing centre to get the job read in. Every now and then a card deck fell down; in lucky cases, the cards carried numbers and could be sorted mechanically. When I started my first MRD-CI calculations in 1976, the maximum amount of RAM a user was allowed to allocate on the IBM 370/168 main frame at the Bonn Computing Center was 1 MB. Typically, it took two weeks of queuing time before job execution. The machine was regularly shut down over night and on weekends, save for an occasional Saturday when our workgroup had exclusive access to the whole machine for performing the most elaborate quantum chemical calculations. I also recall that two-electron integrals – even for a small molecule such as ethylene – had to be stored on magnetic tapes that were rewound after each pass. From this perspective, the high quality of the quantum chemical results achieved by the MRD-CI method in the early days [8] is quite amazing.