Linux Memory Management:
Its similarities and difference with Windows’ Memory Management specifically in
Page swapping algorithms

Abstract
Memory management in all the operating system is important because it is crucial for both programming and system administration. In order to make the system faster is that to use an algorithm that saves the time for managing processes and jobs. The article stated the brief discussion on how the Linux memory management works and thus it states that it works almost the same as what the Windows operating system with 32-bit and 64-bit is using. All i/o Linux goes through the page cache and area that caches to improve disk i/o performance same as with the one used in Windows’ cache and Virtual Memory. The paper discussed the difference and similarities of Windows’ memory management and Linux’s Memory Management. It concludes that Linux’s memory management is simpler and easier to maintain compared to Windows’ though, thus Window’s memory management favors the performance and Linux’s having its operating system open sourced, needs improvement through some researches and open sourced programmers.

Keywords: Linux, Unix, Page Swapping, Windows, Memory, Virtual Memory, Paging, Placement Policy, Least Recently Used (LRU), Not Recently Used (NRU)

Linux Memory Management:
Its similarities and difference with Windows’ Memory Management specifically in
Page swapping algorithms Operating systems is the heart of all computers may it be for personal use, schools and government use. It is the bridge from the end user to the hardware components of the system. It gives the end user the privilege to understand on how the computer works. Memory management is the most important part in the operating system, the faster the computer is the better. Understanding the memory management of the two operating systems namely Linux and Windows is important to be able to differentiate its purpose and thus give the end user knowledge on what is the better OS when it comes to memory.
Literature Review
In Charles D. Cranor and Gurudatta M. Parulkar (1999)’s article reviewing the virtual memory system, The Memory Management System is one of the important part of an operating system. Its function is to manage the memory hierarchy of RAM and hard disks. Its job include allocation and de-allocation of memory to processes of logistics, and the us of Virtual Memory by making use of hard disk as extra for RAM. The Memory system should be optimized, since the performance affects the performance and speed of the system. In C. Cranor. (1998)’ article about the implementation of the uvm virtual memory system, For the virtual memory system to work, mapping functions is required that performs address translation, that converts virtual address to physical address. Virtual address is an address that the applications use to refer to a memory location, and physical address is an actual memory location transferred to the local memory bus. According to Flynn and Mchoes (2011), placement policy is of importance only for optimizing certain behavior. Behavior of a paging system is dependent only on the fetch and placement policy. For the fetch policy a demand paging system is used in which the system brings a page to memory only when needed. For page replacement policy, LRU is used in most of the windows’ operating system.
Discussion
The Linux has been founded by Linux Torvalds. Windows and Linux memory management is similar in some aspects. Data structure and its features are similar. Some similarities of these systems are stated by E. G. Coffman and P. J. Denning (1998) – Hardware Abstraction Layer (HAL) which all of the OS have, that does the system-dependent work, and ables the rest of the kernel to be coded in platform independent. Copy-on-write: According to Dennig and Coffman when a page is to be shared, the system uses only one page with both processes sharing that same copy of the page. But, when one of the processes does a write into the page, a copy is made for that process, which it can then manipulated independently. Due to this characteristic it makes the system efficient. Background daemon: it is invoked periodically and performs tasks like page flushing, freeing unused memory. Shadow paging: A shadow object is created for an original object such that the shadow object has some of its pages modified from the original object, but shares the rest of the pages with the original object. Inter-Process Communication: The memory mapped files are allowed to be then shared between processes forming a method for inter-process communication. Memory mapped Files: A file can be mapped onto memory, can be used with memory read/write instructions.
Data structures used by Windows NT are in a tree form. Each node of the tree is called Virtual Address Descriptors (VAD). Each VAD denotes a range of address which has the same protection parameters and commit state information. It is balanced, which means that depth of the tree is short, which makes the search time shorter. The VAD marks each node as used, free, or reserved.
Linux has a virtual memory data structures as of UNIX. These are structures which represent continuous memory areas which have the same protection parameters. This list is searched whenever a page is to be found at a particular location. The structure records the range of address it is mapping into, protection mode, whether it is not page-able, and the direction it will be in. It records whether it is public or not. If the number of entries is greater than 32 then it is converted in to tree form from link list. Distribution of Process Address Space in Linux, usually 3GB is kept for the process and 1 GB given to the kernel, while in Windows; 2GB are kept for each. Page Replacement is important part in memory management. It is concerned with choosing which page swap out of the memory, whenever there is a need for more free memory. The Optimal Algorithm is the most ideal algorithm for page replacement it removes the page which will be required in the most distant future. It will decrease the number of page faults thus improve the system by minimizing swapping or read and write functions at the same time.
Up to Linux 2.2, the Linux Virtual Memory (VM) had focused on simplicity and low overhead. Hence it was rather quite primitive and had many problems, especially under heavy load. It was influenced by System V. However Riel has worked a lot on the Linux VM in the past couple of years, and improved it a lot for the Linux release. His discussion with Linux uses a demand paged system with no pre-paging. Until kernel version 2.2, Linux used NRU algorithm for page replacement, but due to the various shortcomings of the algorithm, they have changed it and implemented an approximate Least Recently Used in 2.4.The aging to effect LRU is brought about by increasing the age a counter associated with a page of to be referenced during a scan, and, decreased exponentially divided by 2 when found not to have been referenced according to Mathew Dillon and Rik Van Riel (2000)
Limitation
The limitation of this case study is to only compare the performance of the Linux OS with Windows OS regardless of the version but with only general information by its memory management. The topic discussed here is only concerned with the basic algorithm or concept used in improving each one’s page swapping thus resulting to the efficiency of the operating system.
Conclusion
Linux and Windows are both modern when it comes to memory management, both of the uses the LRU algorithm in swapping or similar to LRU concept which is modified when it comes to higher versions of windows. Linux’s memory management is simpler and easier to maintain compared to Windows’ though, thus Window’s memory management favors the performance and Linux’s having its operating system open sourced, needs improvement through some researches and open sourced programmers. The develop Open Source Operating systems, which are maintained by a lot of hackers and programmers around the world, it will be expected that in the future, these operating systems will become better than the Windows operating systems.