In information technology, the term redundant can refer to various situations:
- The term redundant can describe computer or network system components, such as fans, hard disk drives, servers, operating systems, switches and telecommunication links that are installed to back up primary resources in case they fail. A well-known example of a redundant system is the redundant array of independent disks (RAID).
- Redundant information is unneeded or duplicated information.
- Redundant bits are extra binary digits that are generated and moved with a data transfer to ensure that no bits were lost during the data transfer.
- Redundant data can protect a storage array against data loss in the event of a hard disk failure.
Redundancy in storage
When it comes to storage, redundancy can be a safeguard or take the form of unwanted overhead. Data volumes will often contain redundant storage blocks. A deduplication process may remove these redundant blocks to reduce storage consumption within the volume or to minimize the volume of data that must be backed up.
Many organizations may intentionally create redundant copies of data to minimize the chance of data loss. This redundancy may exist as mirrored virtual machines (VMs) or storage volumes, or as an offsite, synchronized data copy.
Data redundancy vs. backup
Data redundancy and backups are both intended to prevent data loss, but the two technologies are quite different. Data redundancy often takes the form of a synchronized copy of the organization’s data. For example, an organization might create a redundant VM or storage volume.
Data redundancy can help to prevent service outages. For example, if a VM were to fail, a replica VM could quickly be brought online to minimize the service disruption. Backups, on the other hand, are copies of data and other resources.
In recent years there has been a lot of overlap between redundancy and backups. However, backups are usually best suited for point-in-time recovery capabilities. If a system needs to revert to a previous state, backups are used. Some data redundancy products also offer point-in-time recovery capabilities, but generally have fewer recovery point options. Backups are also a good choice for granular recovery (recovering a single file, email message and so on). In contrast, redundant systems are better suited to situations in which the organization needs to keep critical systems online and cannot tolerate a long recovery period.
Redundancy in RAID
One of the most common forms of data redundancy is RAID. RAID arrays are designed to provide better performance and/or reliability than what is possible using a single disk.
RAID can refer to a number of different storage architectures, which are known as RAID levels. Not all RAID levels provide data redundancy, but most do. RAID 1, for instance, mirrors disks so an exact copy of the disk can be used if the primary copy fails.
RAID 5 provides redundancy through the use of parity. Data is striped across all the disks in the array in a way that spans less data across multiple disks. Each disk also contains parity information that can keep the array working in the event of a disk failure. When a failed disk is replaced, the parity information is used to reconstruct the contents of the failed disk onto the new disk.
There are many other RAID levels that protect data through redundancy. Parity and mirroring are simply two of the most common examples.