RAID5 with Hotspare versus RAID6 - Starline Computer: Storage und Server Lösungen von erfahrenen Experten

RAID5 with Hotspare versus RAID6

The scenario: A RAID system with six bays can be configured differently. On the one hand as RAID5 with hot spare and on the other hand as RAID6 with all drives under steam.

Both are common methods for improving data security and availability, but they offer different advantages and disadvantages in terms of redundancy and rebuild. In the following, we highlight the differences in each case.

RAID5H vs RAID6

Redundancy and reliability

RAID5 with hot spare:

  • RAID5 offers one parity block per stripe. This means that a drive can fail without data being lost. The data of the failed drive can be restored from the remaining data carriers.
  • The hot spare drive is an additional drive that remains inactive until another drive in the RAID5 fails. In the event of a failure, the hot spare drive automatically takes over the role of the failed drive and the recovery process (rebuild) begins immediately. As long as the rebuild is running, the system is susceptible to another failure, which can lead to data loss.
  • Fail-safe: RAID5 with Hotspare can only cope with one drive failure.

RAID6:

  • RAID6 offers two parity blocks per stripe, which are distributed across the drives. This means that two drives can fail at the same time without data being lost.
  • There is no hot spare drive in RAID6 as the parity blocks themselves provide redundancy. The rebuild starts immediately when the defective drive is replaced.
  • Fail-safe: RAID6 can survive two drive failures before data is jeopardised.

Storage capacity

RAID5 with hot spare:

With 6 drives, the capacity of 4 drives is used for data storage: 5 minus 1 (for parity) minus 1 (hot spare)
 

RAID6:

With 6 drives, the capacity of 4 drives is also used for storage: 6 minus 2 (for parity)

Recovery (Rebuild)

RAID5 with Hotspare:

In the event of a drive failure, the hot spare drive is activated immediately and starts rebuilding the data. As all blocks on all remaining disks have to be read during a rebuild, the probability that sector errors and timeouts will also occur on other disks is greater than in normal operation. If a further disc fails as a result, the RAIDset is corrupted with RAID 5 and the data is no longer available.
 

RAID6:

With RAID6, the admin can first make a backup of the most important data. This is often the latest data that has only recently been written. The probability of sector errors is very low. He can then replace the failed disc and start the rebuild. If a second disc fails during this process, the data remains available with RAID 6.

The risk of data loss during the rebuild is therefore lower, as there is no risk of data loss in the event of another failure (within the two parity blocks).

Conclusion

In an environment where data security is the top priority, RAID6 is clearly the better choice as it can cope with two drive failures. After all, it offers the same net capacity as RAID5 with hot spare and is also much more secure against data loss.

Any questions?

KB
Konrad Beyer
Technical Support

Our technical manager has a comprehensive knowledge of all storage and server topics.