File Systems and Disk Management

File Systems and Disk Management

• Long-term information storage must store large amounts of data, survive process termination, and allow multiple processes to access information concurrently.

File Naming

• Files have typical extensions.

File Structure

• There are three kinds of files: byte sequence, record sequence, and tree.

File Types

• There are various file types, including executable files and archives.

File Access

• Sequential access reads all bytes/records from the beginning and cannot jump around, convenient for magnetic tapes.
• Random access reads bytes/records in any order, essential for database systems.

File Attributes

• Files have possible attributes.

File Operations

• File operations include create, delete, open, close, read, unlink, append, seek, get attributes, set attributes, and rename.

File System Implementation

• A possible file system layout consists of a file allocation table, inode table, and data blocks.
• Contiguous allocation of disk space can lead to fragmentation.
• Linked list allocation uses a file allocation table in RAM.
• Inode allocation uses a file allocation table in RAM and a disk block pointer.

Implementing Directories

• A simple directory has fixed-size entries with disk addresses and attributes.
• Directory entries can refer to an inode instead of containing attributes.
• Handling long file names in directories can be done in-line or using a heap.

Shared Files

• Shared files can be implemented using a common inode.
• Linking a file creates a new directory entry pointing to the same inode.
• Removing a shared file only removes the directory entry, not the file itself.

Disk Space Management

• Block size affects data rate and disk space efficiency.
• Free space can be managed using a linked list or bit map.
• Quotas can be used to track and limit disk use per user.

File System Reliability

• File systems need to be dumped regularly to ensure reliability.
• Bit maps can be used to keep track of modified files and directories.
• File system states can be consistent, missing a block, or have duplicate blocks.

File System Performance

• The block cache can improve file system performance.
• Log-structured file systems use a log of writes to improve performance.

Example File Systems

CD-ROM File Systems

• CD-ROM file systems use ISO 9660 directory entries.

CP/M File System

• CP/M memory layout includes a directory and file allocation table.
• CP/M directory entries have a specific format.

MS-DOS File System

• MS-DOS directory entries have a specific format.
• Maximum partition sizes depend on block size.

Windows 98 File System

• Windows 98 uses an extended MS-DOS directory entry.
• Long file names are stored in a specific format.

UNIX V7 File System

• UNIX V7 directory entries have a specific format.
• UNIX V7 inodes contain file attributes and disk block pointers.
• Looking up a file involves traversing the directory hierarchy.

An Example Program Using File System Calls

• Example programs can use file system calls to create, read, and write files.

Memory-Mapped Files

• Memory-mapped files can be used to map files into a process's address space.

Directories

Single-Level Directory Systems

• Single-level directory systems have a single directory with all files.

Two-level Directory Systems

• Two-level directory systems have a root directory with subdirectories.

Hierarchical Directory Systems

• Hierarchical directory systems have a tree-like structure.

Path Names

• Path names are used to locate files in a directory hierarchy.

Directory Operations

• Directory operations include create, delete, opendir, closedir, readdir, rename, and link.

This quiz covers the basics of file systems, including files, directories, and file system implementation. It also touches on long-term information storage and the requirements for it.