Solid-state disks (SSDs) have a different end-of-life behavior than rotating media (hard disk drives, or HDDs). Data ONTAP monitors and manages SSDs to maximize storage performance and availability.
In the absence of a mechanical failure, rotating media can serve data almost indefinitely. This is not true for SSDs, which can accept only a finite (though very large) number of write operations. SSDs provide a set of internal spare capacity, called spare blocks, that can be used to replace blocks that have reached their write operation limit. After all of the spare blocks have been used, the next block that reaches its limit causes the disk to fail.
Because a drive failure is an undesirable occurrence, Data ONTAP replaces SSDs before they reach their limit. When a predetermined percentage of the spare blocks have been used (approximately 90%), Data ONTAP performs the following actions:
- Sends an AutoSupport message.
- If a spare SSD is available, starts a disk copy to that spare.
- If no spare is available, starts a periodic check for a spare so that the disk copy can be started when a spare becomes available.
- When the disk copy finishes, fails the disk.
Note: You do not need to replace SSDs before they are failed by Data ONTAP. However, when you use SSDs in your storage system (as for all disk types), it is important to ensure that you have sufficient hot spares available at all times.