RAID 50

RAID 50, also known as RAID 5+0, is a type of nested RAID level that combines distributed parity (RAID 5) with striping (RAID 0). It uses the block-level striping of RAID 0, and has better write or I/O capabilities than standard RAID 5 array, along with enhanced fault tolerance.

RAID 50 requires at least six drives to constitute.

RAID 50 offers excellent overall performance and reliable data storage with larger volumes. In this RAID, data can be recovered even if two physical disks (one in each array) fail at the same time. Failure of up to four drive can be overcome as long as each failed drive takes place in a different RAID 5 array. This RAID level is best suited for applications that require high read/write speed, blazing fast data transmission and highly reliable data storage. These applications consist transaction and office applications for small files access of many users.

The Disadvantages Associated to RAID 50 are –

  • It’s very expensive to implement and requires a sophisticated controller.

  • Failure of two or more drives in one of the RAID 5 segments leads to the entire array idle.

  • All disk spindles must be synchronized that limits your choice of drives.

Here are some of the pros and cons of RAID 50:

Disc Space: While RAID 50 may require more overhead space than RAID 5, it needs far less space than RAID 10. This makes it an optimal choice.

Risk: In case of a setup like RAID 5, the failure of a second disc can compromise the entire array. Such an issue gets mitigated with RAID 50 as it can tolerate multiple disc faults, as long as, of course, the faults happen in the right places.

Performance: While RAID 50 does not perform as well as RAID 10 during a rebuild, it does much better than RAID 5 in terms of write performance.