--- title: "Page size impact; internal, external fragmentation" subject: "Operating System" chapter: 4 key: "semester_6/operating_system:page-size-impact-fragmentation" source: https://www.bctengineer.com/q/semester_6/operating_system/page-size-impact-fragmentation site: BCT Engineering --- # Page size impact; internal, external fragmentation Operating System · Chapter 4: Memory Management - Asked in: 2075 Baisakh, 2+2 marks - Video: [Contiguous Memory Allocation ll Internal and External Fragmentation Explained with Example in Hindi](https://www.youtube.com/watch?v=bIpx966TDMo) — 5 Minutes Engineering - Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14 - Page: https://www.bctengineer.com/q/semester_6/operating_system/page-size-impact-fragmentation - Whole chapter: https://www.bctengineer.com/md/semester_6/operating_system/chapter-4 ## Question What is the impact of size of page in virtual memory management performance? Define internal and external fragmentation. ## Short answer Small pages reduce internal fragmentation but increase page-table size and faults; large pages improve I/O efficiency but waste more memory. Internal fragmentation is unused space within an allocated block; external fragmentation is free memory split into small non-contiguous holes. ## Answer #### Impact of page size on performance Page size is a trade-off: | Small pages | Large pages | |---|---| | Less internal fragmentation (average waste = half a page per process) | More internal fragmentation | | Larger page table, more memory for tables | Smaller page table | | More page faults and more TLB misses/entries needed | Fewer faults, better TLB coverage | | Many small disk transfers; more seek/rotation overhead per byte | Efficient disk transfer (one seek per large block) | | Better match to locality; less unused data loaded | May load data that is never used | According to Tanenbaum, for average process size *s* and page-table entry size *e*, the overhead is minimised at page size p = √(2se). Typical sizes are 4 KB to 8 KB. #### Internal fragmentation Wasted memory **inside** an allocated block or partition because the block is larger than the process needs, so the unused part cannot be given to anybody else. Example: a 9 KB process in 4 KB pages needs 3 pages (12 KB) and wastes 3 KB; a 5 KB process in an 8 KB fixed partition wastes 3 KB. It occurs in paging and fixed partitioning. #### External fragmentation Total free memory is enough for a request, but it is split into **small non-contiguous holes**, none large enough. Example: holes of 10 K and 15 K cannot satisfy a 20 K request. It occurs in variable partitioning and segmentation; it is cured by compaction (or avoided by paging). ## Check yourself The answer's key points as multiple-choice questions, with the correct option and why. 1. What is a drawback of small pages? A. More internal fragmentation B. Larger page table and more TLB misses C. Poor match to locality D. Fewer page faults Answer: B. Small pages mean more pages per process, so a bigger page table and more TLB entries needed. (Key point: Small page effect) 2. Which is a drawback of large pages? A. Larger page tables B. Greater internal fragmentation and possibly loading unused data C. More seek overhead per byte D. More page faults Answer: B. Large pages waste more of the last page and may bring in data never used. (Key point: Large page effect) 3. According to Tanenbaum, for average process size s and entry size e, overhead is minimised at which page size? A. p = s*e B. p = sqrt(2se) C. p = 2se D. p = s/e Answer: B. Balancing page-table size and internal waste gives p = sqrt(2se). (Key point: Optimal page size) 4. Which pair is correct? A. Internal: free memory split into small holes; External: waste inside a block B. Internal: waste inside an allocated block; External: enough total free memory but in small non-contiguous holes C. Both are waste inside blocks D. Both arise only in paging Answer: B. Internal fragmentation is inside the block (e.g. 9 KB in 12 KB of pages wastes 3 KB); external is scattered holes. (Key point: Fragmentation types) ## Related in Operating System - [Virtual memory management; thrashing](https://www.bctengineer.com/question.md?key=semester_6%2Foperating_system%3Avirtual-memory-thrashing) — Operating System, chapter 4 - [Page fault; virtual to physical address](https://www.bctengineer.com/question.md?key=semester_6%2Foperating_system%3Apage-fault-64k-virtual-32k-physical) — Operating System, chapter 4 - [Optimal replacement; fragmentation, utilization](https://www.bctengineer.com/question.md?key=semester_6%2Foperating_system%3Aoptimal-fragmentation-utilization) — Operating System, chapter 4 - [File system vs virtual memory; contiguous allocation](https://www.bctengineer.com/question.md?key=semester_6%2Foperating_system%3Afs-vs-virtual-memory-contiguous-tape) — Operating System, chapter 5 - [Segmentation; page table size; TLB](https://www.bctengineer.com/question.md?key=semester_6%2Foperating_system%3Asegmentation-address-bits-tlb) — Operating System, chapter 4 - [Linked list file system; block size impact](https://www.bctengineer.com/question.md?key=semester_6%2Foperating_system%3Afs-linked-list-block-size) — Operating System, chapter 5 ## Related in other subjects - [Risk planning; internal, external; risk reduction](https://www.bctengineer.com/question.md?key=semester_7%2Fproject_engineering%3Arisk-reduction-internal-external-planning) — Project Engineering, chapter 5 - [Noise, distortion, interference; internal and external noise](https://www.bctengineer.com/question.md?key=semester_6%2Fcommunication_systems%3Anoise-types-internal-external) — Communication Systems, chapter 1 - [External vs internal noise; interference sources; need for modulation](https://www.bctengineer.com/question.md?key=semester_6%2Fcommunication_systems%3Aexternal-internal-noise-interference-needs) — Communication Systems, chapter 1 - [Noise; internal vs external; sources of external noise](https://www.bctengineer.com/question.md?key=semester_6%2Fcommunication_systems%3Anoise-internal-external-sources) — Communication Systems, chapter 1