Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] abuf_v2_0.gz(20 Kbytes)
Manuscript Title: OLYMPUS and preprocessor package for an IBM 370/165.
Authors: M.H. Hughes, K.V. Roberts, P.D. Roberts
Program title: OLYMPUS FOR IBM 370/165
Catalogue identifier: ABUF_v2_0
Distribution format: gz
Journal reference: Comput. Phys. Commun. 9(1975)51
Programming language: Fortran.
Computer: IBM 370/165.
Operating system: HASP.
RAM: 72K words
Word size: 8
Peripherals: disc.
Keywords: General purpose, Preprocessor, Control, Utility, Package, Initial-value problems, Simulation, Computation, Standard, Framework, Kernel, Programming practice.
Classification: 4.14, 20.

Nature of problem:
A standard methodology called the OLYMPUS system has been established for constructing, testing and operating Fortran programs which solve equations describing initial-value problems. A previously-published control and utility package implements this system for the ICL 4/70, and the present package contains the corresponding version for IBM 360/370 series computers. It also includes a Fortran preprocessor which will insert labelled COMMON blocks into a source file using only one master copy of COMMON.

Solution method:
Comp. Phys. Commun. 7(1974)245. should be consulted for the details of the OLYMPUS system proper. The preprocessor reads the COMMON blocks into store on one channel NA, the source code on another channel NB, and outputs the expanded program as a data set on a third channel NC which is then passed to the compiler.

Different implementations of OLYMPUS are required for different types of computer to take account of differing wordlengths, channel numbers etc. The present package should work on most IBM 360 and 370 systems provided that the control codes are modified to suit local conventions. Since the preprocessor is written in Standard FORTRAN it can, in principle, be used on any computer. However, it has been designed to take advantage of the IBM scheme for concatenating data sets.