File Systems and Structure

Study Notes

File systems provide a way to store and retrieve data and programs from storage media
Files are collections of related information defined by the creator
They are mapped to disks or other storage media by the operating system
File systems create a convenient way to write, read, save, and retrieve data
A file is a collection of related information mapped to secondary storage
Information in files is stored as bits, bytes, lines, or records
The user only sees a logical view of the file
The system handles the mapping of logical files to secondary storage

File Management: Manages how files are stored, referenced, shared, and secured
File Allocation: Provides methods to allocate files on disk space
File Access Methods: Provides methods to access stored files

A file name has two parts separated by a period (.)
The first part is the file name defined by the user
The second part is the extension, which indicates the file type
Operating systems need to recognize file types to perform appropriate operations
Examples of file type extensions:
- Source code: .c
- Object file: .obj or .o
- Executable file: .exe, .com, or .bin
- Text file: .txt or .doc
- Batch file: .bat
- Archive file: .zip or .rar
- Multimedia file: .jpg, .mp3, or .mov

Sequential File Access: Simplest method. Information is processed in order, one record at a time. Common for editors and compilers
Direct File Access: Also called relative access. Accesses records directly by their position, allowing rapid reading and writing in any order.
Indexed Sequential File Access: Uses an index to provide faster access to records in a sequential file. Similar to an index in a book that points to specific pages.

Disk scheduling is important for I/O operations for reasons such as:
- Multiple processes may send I/O requests, but only one can be serviced at a time.
- When disk storage is close to full.
- Random user requests may require scheduling.

FCFS (First Come First Serve): Processes requests in the order they arrive. Simple, but can be inefficient.
SSTF (Shortest Seek Time First): Selects the request closest to the current disk head position, minimizing seek time, but potential for starvation of requests farther out.
SCAN (Elevator Algorithm): Treats the disk arm as an elevator. Services requests in one direction until the end, then reverses and services requests in the other direction. Prevents starvation.
LOOK: Similar to SCAN, but reverses direction only when there are no further requests in the current direction. More efficient than SCAN.
C-LOOK (Circular LOOK): Variant of LOOK that does not reverse direction at the ends but treats the disk as circular, continuing to serve requests in the current direction.