File system

A file system is a set of rules that manage the way files are stored and retrieved on a disk. It defines how data is written to disk.

Cluster - Allocation unit

This is the smallest allocation unit imposed by the file system. It is made up of multiple sectors (the smallest physical part: ~512 bytes). The cluster is the smallest space a file can occupy on a disk.

For example: if the cluster size is 4 Kb, to store a 10 Kb file, 3 clusters are needed (for a total of 12 Kb). The remaining 2 Kb is lost. The space lost is included in the file size.

External disk

• FAT32

  • Compatible with all major operating systems
  • Maximum file size is 4 GB
  • Minimum allocation size is 512 bytes

• ExFAT (enhanced FAT32)

  • No file size limit
  • Minimum file size for allocation is 4 Kb

• NTFS

  • Proprietary file system developed by Windows
  • Free drivers available for Linux
  • No file size limit
  • Minimum allocation size is 512 bytes (for small volumes)

Default cluster size for NTFS, FAT, and exFAT

RAID

Following RAID definitions come from www.sobyte.net

RAID 0: Bandwidth groups without error control

The speed of RAID 0 is the fastest of all levels. But RAID 0 has no redundancy, if one disk (physical) is damaged, all data is unusable. More than two hard drives are required to implement RAID 0. RAID 0 implements striped groups where data is not stored on one hard drive, but is divided into data blocks that are stored on different drives. It is most efficient if just the required data is on different drives. It does not need to calculate checksums and is easy to implement. Its disadvantage is that it does not have data error control, if the data in one drive is incorrect, even if the data on other drives is correct, it will not help.

RAID 1: Mirror image structure

The mirror hard disk is equivalent to a backup disk, so you can imagine that the security of this hard disk mode is very high, RAID 1 data security is the best in all RAID levels. However, its disk utilization is only 50%, which is the lowest of all RAID levels.

RAID 2: with Hemming code checksum

RAID 2 uses certain encoding techniques to provide error checking and recovery. This encoding technique requires multiple disks to hold the inspection and recovery information, making RAID 2 technology more complex to implement.

RAID 3: Parallel transfer with parity code

RAID 3 uses a single disk to store parity information. If a disk fails, the parity disk and other data disks can regenerate the data. RAID 3 provides good transfer rates for large amounts of sequential data, but for random data, the parity disk can become a bottleneck for write operations. Using separate parity disks to protect data is not as secure as mirroring, but the hard disk utilization is greatly improved to (n-1)/n.

RAID 4: Independent disk structure with parity

Accesses data on a block-by-block basis, that is, disk-by-disk, one disk at a time. it is much more difficult to recover from a failure than RAID 3.

RAID 5: Distributed Parity Independent Disk Architecture

The biggest advantage of RAID 5 is that if a disk drops, the RAID will work as usual. RAID 5 has high read efficiency, average write efficiency, and good block collective access efficiency. Because the parity code is on a different disk, it improves reliability and allows for individual disk errors. Any hard drive that is corrupted can reconstruct the corrupted data based on the checksum bits on the other hard drives. The hard disk utilization is n-1, but it does not address data transfer parallelism well, and the controller design is quite difficult.

RAID6: Two disk structures with parity codes for storage

It is an extension of RAID 5 and is mainly used in situations that require data to be absolutely error-free.

NAS hard drives

RAID Calculator

Difference between Pro & non-pro:

• Pro version -- 5 year-warranty
• Non pro version -- 3 year-warranty