Introduction (Click to View)
Reliable Information - Notice (Click to View)
RAID Volumes & Storage Devices
An Introduction to RAID.
With RAID (Redundant Array of Independent Disks), many disks can act as a backups for each other to increase reliability, performance and fault tolerance.
NOTE: RAID can also stand for "Redundant Array of Inexpensive Disks" And the "D" can stand for "Devices".
RAID Levels
The RAID advisory board defines RAID levels. The most common levels are numbered from 0 to 6 where each level corresponds to a specific type of fault tolerance. Only levels 0, 1, and 5 are of much relevance to the desktop, so that's what we will be covering. See table below:
Raid 0 - Striping Without Parity
Disk striping is a technique where data is divided into 64k blocks and spread(divided) into a fixed order across all the disks in the RAID array. Because it provides no redundancy , if any disk in the array fails all data is lost. However RAID 0 is often used to improve performance on mission critical machines as it spreads disk input/output over multiple drives.
RAID 0 requires a minimum of 2 disks and can handle a maximum of 32. It provides the best performance over other RAID levels when spread across multiple disk controllers.
Raid 1 - Mirroring/Duplexing
Mirroring requires an even number of disks as well as a single disk controller. RAID 1 works at the partition level so all partitions (including boot/system partitions) can be copied or mirrored.
Each action is mirrored by the second disk in the set, it's basically a back up of the entire drive. RAID 1 is the simplest way of protecting against data lost on disks. If one disks fails, the copy takes over and then you just need to insert a new secondary disk for RAID 1 to start mirroring again.
NOTE: Duplexing is simply a RAID 1 volume with an extra disk controller on the second drive, this helps improve performance and reduce channel traffic. Duplexing is designed to protect against controller failures as well. RAID 1 requires only 2 disks.
Raid 5 - Striping with parity
RAID 5 is the most common for fault tolerant machines. It's different from other levels in that it writes parity information across all disks in the array.
The parity is structured so that enough information is spread across the array that if one disk fails there is enough information to the other disks to detect and correct the errors by completely reconstructing the lost data.
RAID 5 offers the best performance when it comes to read operations, however when a disk fails the read performance as it is degraded by the need to calculate and re-write data loss. Normal operations do require almost three times more memory due to parity calculation.
RAID 5 requires a minimum of three disks.
This concludes my information guide of RAID volumes, please post if you have any questions or would like to know more, it's interesting stuff and not that complicated once you understand the principles.
Thank you this is Information Guide 1.