File Systems

Explain the inode approach to file allocation with its advantages and disadvantages.

An inode stores information about a file, including its user and group IDs, type, creation/access/modification information, inode modification information, link count, size, and disk-block addresses. The original inode diagram is preserved below.

Advantages: Only inodes for open files need to be in memory, so the memory requirement is comparatively small. The approach also supports large files efficiently.

Disadvantages: Inodes and data blocks may be far apart, causing a long initial seek. Inodes for files in the same directory may not be adjacent, making searches less efficient.

Explain file-allocation and file-access methods, with their advantages and disadvantages. The exam heading is retained in the original scan.

Allocation methods

• Contiguous allocation: Stores a file in consecutive blocks. For example, a 50 KB file uses fifty consecutive 1 KB blocks. Sequential and direct access are simple, but fragmentation and difficulty growing a file can waste space.
• Linked allocation: Each block holds a pointer to the next block and the remaining space holds data. Files can grow without requiring adjacent blocks. Random access is slow because the links must be followed.
• Linked allocation with an index table: Moves the block pointers into a table kept in memory. This makes following the chain faster, but requires memory for the table.
• Inode/indexed allocation: Uses a file's index information to locate blocks. It supports random access and avoids external fragmentation, with pointer-storage overhead and possible wasted index space.

Access methods

• Sequential access: Processes records in order. It is simple, but reaching a later record requires passing earlier records.
• Direct access: Accesses a selected record without reading all preceding records. The source describes selection by a key rather than sequential position.

Explain similarities between a file system and virtual memory. Which access method is best for tape?

Virtual memory presents an apparently contiguous address space while physical frames may be scattered. Translation and backing storage allow a process's address space to grow beyond physical memory. Similarly, a file system presents a continuous file even when its disk blocks are not contiguous; the file can grow without moving all of its existing contents.

Sequential access is suitable for tape: records are read in order, advancing the read position.

Explain block caching, read-ahead, and reducing disk-arm motion as methods of improving file-system performance.

• Block caching: Keeps frequently used disk blocks in memory. First look for a requested block in the cache; if absent, read it from disk and place it in the cache.
• Read-ahead: When sequential access is expected, fetch the next block before it is requested. The source illustrates this as reading block k + 1 after block k.
• Reducing disk-arm motion: Place blocks likely to be accessed together near one another, preferably in the same cylinder. The source discusses the separate seeks required to access an inode and its data and the cost of locating inodes near one end of the disk.

The original notation and full discussion are retained here:

Write a short note on the UNIX file system.

The source describes a file system descended from Version 7 UNIX. Its layout includes boot blocks, a superblock, and cylinder groups. The superblock holds identifying information such as a magic number, geometry, and statistics. See the original answer above for the handwritten details.

Explain file attributes, single-level and hierarchical directories, and methods of managing free disk blocks.

File attributes describe a file and control settings such as access rights. A single-level directory keeps all files in one directory. A hierarchical directory organizes files within directories and subdirectories, providing a logical grouping.

Free-space methods listed in the source are:

1. Bitmap: One bit represents each block. The source uses 0 for allocated and 1 for free.
2. Linked list: Links the free blocks; the first pointer can be kept in memory to locate the list.
3. Grouping: Listed as a free-space method, but the handwritten explanation is incomplete and overlaps with linked-list terminology.
4. Counting: Records the first block of a contiguous free region together with the number of free blocks in that region.

Original directory diagrams, bitmap, and free-space notation:

Explain the layout of a file system.

The source places the master boot record in sector 0, with a partition table near its end and one partition marked active. During boot, the BIOS executes the MBR code, which identifies the active partition and reads its boot block. The boot block then loads the operating system. The original disk-layout diagram is retained below.

What is a file?

A file is a collection of related information defined by its creator. It is an abstraction through which the kernel organizes information on nonvolatile storage, such as hard disks, floppy disks, CDs, and optical media.

Explain file-system layout and differentiate the file-system interface from its implementation.

The layout discussion is given in Question 7. Implementation concerns how files are stored, how space is managed, and how storage is made efficient and reliable. The source lists directory structures, allocation methods such as contiguous, linked, and indexed allocation, and free-space management.

The interface exposes operations such as open, read, write, and seek. It covers access methods, directories, mounting, sharing, and protection. The source also discusses transferring data between memory and secondary storage.