CSC 414 Lecture, Spring 2003

Operating Systems File Systems Outline  Files   Directories  Disk space management  Misc File Systems  Abstraction to disk (convenience) – “The only thing friendly about a disk is that it has persistent storage.” – Devices may be different: tape, IDE/SCSI, NFS  Users – don’t care about detail – care about interface  OS – cares about implementation (efficiency) File System Concepts  Files - store the data  Directories - organize files  Partitions - separate collections of directories (also called “volumes”) – all directory information kept in partition – mount file system to access  Protection - allow/restrict access for files, directories, partitions Files: The User’s Point of View  Naming: how do I refer to it? – blah, BLAH, Blah – file.c, file.com  API: how do I access it? – open() – read() – write() Example: Unix open() int open(char *path, int flags [, int mode])  path is name of file  flags is bitmap to set switch – O_RDONLY, O_WRONLY… – O_CREATE then use mode for perms  success, returns index Unix open() - Under the Hood int fid = open(“blah”, flags); read(fid, …); User Space System Space stdin stdout stderr ... 0 1 2 3 File Structure ... ... File Descriptor (where blocks are) (attributes) (index) File System Implementation Process Descriptor Open File Pointer Array Open File Table File Descriptor Table (in memory copy, one per device) (per process) Disk File sys info File descriptors Copy fd to mem Directories Data Next up: file descriptors! File System Implementation  Which blocks with which file?  File descriptors: – Linked List – Linked List with Index – I-nodes File Descriptor Linked List Allocation  Keep a linked list with disk blocks  Good: – Easy: remember 1 number (location) – Efficient: no space lost in fragmentation  Bad: – Slow: random access bad File Block 0 File Block 1 File Block 2 Physical Block null 4 7 2 File Block 0 File Block 1 null 6 3 Linked List Allocation with Index  Table in memory – faster random access – can be large!  1k blocks, 500K disk  = 2MB! – MS-DOS FAT Physical Block 0 1 null 2 null 3 7 4 5 3 6 2 7 I-nodes  Fast for small files  Can hold big files  Size? – 4 kbyte block Disk Addresses (data blocks) i-node attributes single indirect block double indirect block triple indirect block Outline  Files   Directories   Disk space management  Misc Directories  Just like files, only have special bit set so you cannot modify them  Data in directory Maps File name to File descriptor  Tree structure directory the most flexible – aliases allow files to appear at more than one location Directories  Before reading file, must be opened  Directory entry provides information to get blocks – disk location (block, address) – i-node number  Map ascii name to the file descriptor Hierarchical Directory (Unix)  Tree  Entry: – name – inode number  example: /usr/bob/mbox inode name Hierarchical Directory (Win/FAT)  Tree  Entry: – name - date – type (extension) - block number (w/FAT) – time name type attrib time date block size Outline  Files   Directories   Disk space management   Misc Disk Space Management  n bytes – contiguous – blocks  Similarities with memory management – contiguous is like segmentation  but moving on disk very slow!  so use blocks – blocks are like paging  how to choose block size? Choosing Block Size  Large blocks – wasted space (internal fragmentation)  Small blocks – more seek time since more blocks Data Rate Disk Space Utilization Block size Keeping Track of Free Blocks  Two methods – linked list of disk blocks  one per block or many per block – bitmap of disk blocks  Linked List of Free Blocks (many per block) – 1K block, 16 bit disk block number  = 511 free blocks/block  200 MB disk needs 400 blocks = 400k  Bit Map  200 MB disk needs 20 Mbits  30 blocks = 30 K  1 bit vs. 16 bits (note, these are stored on the disk) Tradeoffs  Only if the disk is nearly full does linked list scheme require fewer blocks  If enough RAM, bitmap method preferred  If only 1 “block” of RAM, and disk is full, bitmap method may be inefficient since have to load multiple blocks – linked list can take first in line File System Performance  Disk access 100,000x slower than memory – reduce number of disk accesses needed!  Block/buffer cache – cache to memory  Full cache? FIFO, LRU, 2nd chance … – exact LRU can be done  LRU inappropriate sometimes – crash w/i-node can lead to inconsistent state – some rarely referenced (double indirect block) Outline  Files   Directories  Disk space management   Misc  – partitions (fdisk, mount) – maintenance – quotas – Linux – WinNT Partitions  mount, unmount – load “super-block” – pick “access point” in file-system  Super-block – file system type – block Size – free blocks – free inodes / usr home tmp Partitions: fdisk  Partition is large group of sectors allocated for a specific purpose – IDE disks limited to 4 physical partitions – logical partition inside physical partition  Specify number of sectors to use  Specify type – magic number recognized by OS File System Maintenance  Format: – create file system structure: super block, inodes – format (Win), mke2fs (Linux)  “Bad blocks” – most disks have some – scandisk (Win) or badblocks (Linux) – add to “bad-blocks” list (file system can ignore)  Defragment – arrange blocks efficiently  Scanning (when system crashes) – lost+found, correcting file descriptors... Linux Filesystem: ext2fs  “Extended (from minix) file system vers 2”  Uses inodes – mode for file, directory, symbolic link ... Linux filesystem: blocks  Default is 1 Kb blocks – small!  For higher performance – performs I/O in chunks (reduce requests) – clusters adjacent requests (block groups)  Group has: – bit-map of free blocks and inodes – copy of super block Linux Filesystem: directories  Special file with names and inodes