Example 9.4.1. Hot, Cold, & Warm.
One typical way to implement redundancy is through the use of hot, cold, and warm sites.
A hot site is a secondary location that is live and replicating in real-time what is happening in production. In the case of the primary site going down, a hot site can failover immediately.
A cold site is a secondary location without equipment. A cold site will take some time to set up and configure in the case of an outage.
A warm site is a secondary location with all equipment and connectivity. The equipment will still need to be turned on and made production ready, but it will not take as long to failover to a warm site as a cold one.
Example 9.4.2. RAID.
RAID is an interesting case of redundancy that occurs at the server storage level. RAID stands for Redundant Array of Inexpensive/Independent Disks and as the name states it uses multiple disks to make reads/writes faster
and to be able to recover if one of the disks fails. It is important to note that RAID is not a backup. Backups are meant to aid in recovery and can be co-located. A RAID array is mean to work on a single machine and help mitigate damages cause by disk failures.
RAID has multiple levels, each of which prioritizes a different aspect:
-
RAID 0: Data is stripped across multiple disks to make reads/writes faster. If a single disk is lost
the whole array goes down.
-
RAID 1: Data is mirrored across multiple disks for redundancy. If a single disk is lost the array can be recovered from the other disks.
-
RAID 5: At least three disks are used in a stripped and mirrored fashion such that read/write speeds are increased
and if a
single disk goes down the array can be rebuilt.
-
RAID 10: A combination of RAID0 and RAID1.