Memory management is a key subject under disciplines of computer science, software engineering and information technology. It is a function of the operating system that is concerned with managing the primary memory of a computer system. With memory management the computer manages every memory allocation to each and every process in the computer system and also manages freeing up of this memory that is allocated. Memory management will keep track of what tasks to allocate memory and what time to allocate to the tasks and what tasks to free up when memory becomes scarce. It will update itself of all changes in the computer memory in relation to tasks or processes.
The protections provided by memory management to the host computer are two; a base register and a limit register. What the base register means is that it holds memory addresses of the smallest legal physical memory whereas the limit register will specify the size of the range. Putting this into a working example we find that if a base register has a value of 300000 and the limit register has 1209000 then a program in this computer system can access memory addresses from the base register value all through to the limit.
Memory management is not complete without talking about dynamic linking, dynamic loading, logical vs. physical address space, swapping, memory allocation, fragmentation, paging and segmentation. Dynamic loading is the process by which routines and calls of a program are not loaded into memory until when the program needs them which means that all the routines and program data are kept on the disk. Routines stored on disk are loaded on request but only the main routine is loaded and executed. The process of linking involves collecting and then combining various modules and pieces of code and data that will be loaded into memory and executed. Operating systems have the ability to load system level libraries to an executing program.