M8 - File Systems (Part 2)

T/F: The original FAT file system was designed by Marc McDonald.

True

T/F: In the FAT file system, the File Allocation Table is dynamically allocated during runtime.

False

T/F: The FAT32 file system uses only 8-bit addresses to identify data blocks.

False

T/F: Long file names in the FAT system are handled by Virtual FAT (VFAT), which acts as an extension to the FAT file system.

True

T/F: In the UNIX Version 7 directory entry, file names can be up to 14 characters and must avoid the "/" character.

True

T/F: A larger block size typically improves data transfer rates but may lead to increased disk fragmentation.

True

T/F: Using smaller block sizes in a file system increases disk utilization by reducing the amount of space wasted.

True

T/F: In file systems, a bitmap can be used to keep track of free blocks, with each bit representing the status of a block.

True

T/F: Power loss during file operations can lead to file system inconsistency.

True

T/F: With a 1 KB block size, a 12-bit address allows up to 4 MB of data in a partition.

True

T/F: File system backups are essential for recovering from hardware failures, human mistakes, and malicious attacks.

True

T/F: Logical backups in file systems copy every file on the disk regardless of changes since the last backup.

False

T/F: Mirrored storage duplicates every write operation on two drives, allowing automatic failover if one disk fails.

True

T/F: Journaling in file systems records changes before finalizing them, which helps in recovering data after unexpected shutdowns.

True

T/F: RAID 6 can handle up to two simultaneous disk failures without data loss due to its dual-parity blocks.

True

How does FAT handle long file names in Virtual FAT?

In a virtual FAT file system, long file names are handled through an extension called "VFAT" (Virtual FAT), where the longer names are stored as hidden directory entries that appear like volume labels, allowing for filenames exceeding the standard 8.3 character limit while still maintaining compatibility with older systems that only recognize short name

What are the Joliet and Rock Ridge extensions for ISO 9660?

Joliet Extensions:

- Enable long file names.

- Support Unicode character sets for special and foreign symbols.

- Allow up to 8 levels of nested directories and additional metadata like timestamps.

Rock Ridge Extensions:

- Extend ISO 9660 to include POSIX metadata for better compatibility with POSIX systems.

What does it mean to have better data rate?

- The disk can read more data per data request

- If block size bigger than avg. file size, blocks may be underutilized, wasting disk space

What are the trade-offs of block size in file systems?

Larger block sizes improve data transfer rates but may lead to worse disk utilization and fragmentation.

Smaller block sizes improve disk utilization but may result in lower data rates as multiple reads might be necessary for small files.

How do file systems track free blocks?

Linked Lists: Dynamic, stored on disk, but can cause frequent memory-disk I/O issues.

Bitmaps: Static size, uses 0s to indicate free blocks and 1s for used blocks, & is simplistic but inefficient

What are common issues with file system consistency after crashes, and their solutions?

- Missing Block: Add the block to the free block list.

- Duplicate Block in Free List: Rebuild the free list.

- Duplicate Data Block: Allocate a new block, copy data, and issue a warning to the user.

Block Consistency

- The correctness of free/allocated block info

- 2 tables are needed (free/allocated)

- Each slot in the table is a counter

What are the differences between block caching in Disk Operating Systems (DOS) and Unix systems?

DOS (Write-Through Cache): Writes modified blocks immediately to disk for consistency, ideal for removable disks.

Unix (Deferred Writing): Delays writing to disk for better performance but requires a sync command for consistency, better for hard disks.

Why is file system reliability critical, and what are some common causes of data loss?

File system reliability is critical because even small data losses can lead to significant financial or operational impacts. Common causes of data loss include hardware failure, malicious attacks, and, most frequently, user mistakes.

Backup Issues

1) Backup all or part of the system

2) Don't back up a file if it has not been changed

3) Compression of backup or not?

4) Difficulty of backup while file system is active

5) Physical security of backup media

What are the types of backups, and what challenges do they present?

Backups can be logical (selective, based on file changes) or physical (block-level duplication). Challenges include determining what to back up, when, and how; managing compressed backups (which slow recovery); and ensuring the physical security of backup media, often through geographic redundancy.

How do RAID levels improve data storage reliability, and what are their trade-offs?

RAID levels improve reliability through redundancy (e.g., RAID 1 mirrors data, RAID 5/6 use parity for fault tolerance). Trade-offs include storage costs (RAID 1), slower write speeds (RAID 5/6), and limited use cases for certain configurations like RAID 2 and RAID 3.

Physical Dumps

- Copies everything, exactly as it is stored on disk starting from block 0

- Very simple to implement

- Runs very fast

- What about bad or unused blocks?

Logical Dumps

- Starts at a directory & traverses the filesystem

- Dumps all files which are changed since the last backup

- Used by most UNIX systems

Mirrored Storage

- Every write is duplicated (write to disk 0 and disk 1)

- Reads are to whichever system is least loaded

- Automatic failover when disk, FS, or network fails

Failover

- Backup operational mode that automatically switches to a standby database, server, or network if the primary system fails

- Costly as the storage space is duplicated

Redundant Arrays of Independent Disks (RAID)

- A way of storing the same data in different places on multiple hard disks or SSDs to protect data in case of a drive failure

- Provide performance gains through parallel hardware access

- Provide reliability through redundant storage

- Fast recovery (not always)

- Hot swap is possible

- Can be software or hardware based

Hot Swap

The removal and replacement of a component in a

system while the power is still on and the system is functioning

RAID 0 (Data Stripping)

- Split files into blocks and scatters them across your physical storage units

- No redundancy but higher performance

RAID 1 (Data mirroring)

- Mirrors data between two drives within the RAID

- Provides data redundancy

- Disk failure tolerance

- Higher storage costs

RAID 5/6

- Uses parity for fault tolerance

- Slower write speeds

What is a journaling file system, and how does it enhance reliability?

A journaling file system logs changes before they are committed, reducing corruption risks after crashes or power failures. It can track either metadata alone or both data and metadata, balancing performance and reliability.