Shifting wearout of storage disks
View Patent ↗Technical solutions are described for shifting wearout of an array of storage disks in a storage system. In an aspect, a method includes staggering completion of maximum number of writes for each storage disk in the array. The method also includes copying data from a storage disk in the array to a replacement disk and redirecting a request to access the data from the storage disk to the replacement disk. The method also includes, in response to passage of a predetermined duration of time since copying the data from the storage disk to the replacement disk, copying the data from the replacement disk to the storage disk and directing the request to access the data from the storage disk back to the storage disk.
1. A method for shifting wearout of storage disks in an array of storage disks in a storage system, the method comprising:
staggering, by a controller of the storage system, completion of a maximum number of writes for each storage disk in the array, the staggering including:
copying, by a controller of the storage system, data from a first storage disk in the array to a replacement disk;
redirecting, by the controller, a first request to access the data from the first storage disk to the replacement disk;
in response to passage of a first predetermined duration of time since copying the data from the first storage disk to the replacement disk, copying, by the controller, the data from the replacement disk to the first storage disk and directing the first request to access the data from the first storage disk back to the first storage disk;
copying, by the controller of the storage system, data from a second storage disk in the array to the replacement disk;
redirecting, by the controller, a second request to access the data from the second storage disk to the replacement disk; and
in response to passage of a second predetermined duration of time since copying the data from the second storage disk to the replacement disk, copying, by the controller, the data from the replacement disk to the second storage disk and directing the second request to access the data from the second storage disk back to the second storage disk, the first predetermined duration and the second predetermined duration are distinct, and the first and the second durations are based on a number of storage disks in the array, and a mean time to failure of the storage disks in the array.
2. The method of claim 1 , wherein the replacement disk comprises a plurality of replacement disks.
3. The method of claim 2 , wherein the data from the first storage disk is copied to a first replacement disk from the plurality of replacement disks, and the data from the second storage disk is copied to a second replacement disk from the plurality of replacement disks.
4. The method of claim 3 , wherein the data from the first storage disk is copied to the first replacement disk in parallel to copying the data from the second storage disk to the second replacement disk.
5. The method of claim 4 , further comprising:
in response to completion of copying data from the first replacement disk to the first storage disk:
copying, by the controller, data from a third storage disk of the array to the first replacement disk;
redirecting, by the controller, a third request to access the data from the third storage disk to the first replacement disk; and
in response to passage of a third predetermined duration of time since copying the data from the third storage disk to the first replacement disk, copying, by the controller, the data from the first replacement disk to the third storage disk and directing the third request to access the data from the third storage disk back to the third storage disk.
6. The method of claim 1 , wherein a storage disk in the array is one of a solid state drive, a flash memory, or a flash dual in-line memory module (DIMM), interfaced using a peripheral component interconnect (PCI) card or a serial attached small computer system interface (SAS).
7. The method of claim 6 , wherein the array of storage disks is maintained as a redundant array of independent disks (RAID).