To begin, memory management is the task carried out by the operating system to accommodate multiple processes by sub-dividing the memory. In a multiprogramming system, proper memory management is a necessity. While the processor is at idle, it is necessary to have memory allocated for other processes to use any remaining processor time. There are five requirements that memory management is meant to handle: Relocation, Protection, Sharing, Logical Organization, and Physical Organization (Stallings, 2012). To help maximize processor utilization, we create a large pool for processes that are ready to be executed to be swapped into from the main memory. When a program is swapped back in, it helps to relocate it to a different part of memory, rather than the same main memory region as before. This is more efficient than to have to find and place the program into the same region. The memory management unit also has to work with concurrency (Memory Management, n.d.). The unit should move the program and deal with memory references and addresses of the program. This will make sure that they make it to the right location in the memory. To help ensure smooth and efficient program usage, each process needs to be protected from interruptions, whether they are intended or by accident (Stallings, 2012). To help prevent these interruptions, any other program will need permission to reference memory locations for reading or writing purposes. We cannot ensure complete protection at compile time because we do not know what the location of the program will be moved to is. To help ensure that memory references refer to memory space given for a specific process, they are checked at run time. Because the operating system is unable to know what memory references a program will make, memory protection is handled by the processor (Stallings, 2012). In order to be more efficient, it is necessary to allow multiple processes to be able to access the same area of the main memory; share the memory. It does not make sense for each process to need its own copy of a program, so they have the ability to share the same copy (Stallings, 2012). Not only does this save room, but time as well. When it comes to inter-process communication, shared memory is one of the fastest techniques used (Memory Management, n.d.) Main memory and secondary memory are organized in a linear or logical fashion. It consists of an order of words. The majority of programs are sorted into modules. Some of these modules can be modified, whereas some of them cannot be modified. There are numerous advantages of an operating system being able to work with programs in the form of modules. Modules can be written by independently, allowing the system to handle the references at run time. Each module would be able to have its own level of protection, based on the information involved. Sharing permissions would be provided on a module level allowing a person to specify what is needed (Stallings, 2012). The way that the memory is organized is called physical organization. The memory is split into two levels: main memory and secondary memory. While the main memory allows for fast access to programs, it does not provide for a permanent storage for the program. This is where the secondary memory comes into play. It is cheaper, provides longer term storage, but it is also significantly slower than the main memory (Stallings, 2012). It is impractical for the user to manage the flow of information between the main memory and the secondary memory. Not only does the user not know where the space in the memory will be to use, but also even with specialized tools, it would be time consuming. Therefore, this is a process that the system should handle (Stallings, 2012).

Works Cited
Memory Management. (n.d.). Retrieved from www.princeton.edu: http://www.princeton.edu/~achaney/tmve/wiki100k/docs/Memory_management.html
Stallings, W. (2012). Operating Systems.