Program dismemberment

Abstract

When we have a large program, we know that it will be swapped in and out of memory during its period of execution. If we have implemented a program using inversion, we also know that it will not be possible for the whole program to be executed (even if it is granted enough time) because there will be points in the code where control will need to be passed between the main program and the subprogram. Loading the program for it to execute also takes time, and it would therefore be logical only to load that part of the program that we know could be executed at any one time. This has the benefit of loading only that code which is required, as well as reducing the amount of time spent loading, which could be better employed in processing the data.