Increased data protection by recovering data from partially-failed solid-state devices
A storage system includes a central storage controller and a solid-state storage device operatively coupled to the central storage controller, the solid-state storage device including a processing device, the processing device to determine whether a die of the solid-state storage device is likely to fail. In response to determining that the die of the solid-state storage device is likely to fail, the processing device is further to mark the die of the solid-state storage device as likely to fail and transmit, to the central storage controller, an indication that the die of the solid-state storage device has been marked as likely to fail.
1. A storage system comprising:
a storage system controller; and
a plurality of solid-state storage devices operatively coupled to the storage system controller, each solid-state storage device comprising a respective processing device and a respective plurality of dies, wherein a processing device of a solid-state storage device of the plurality of solid-state storage device is configured to:
determine whether a die that is one of a plurality of dies of the solid-state storage device is likely to fail based on a number of failed input/output (I/O) operations performed on the die satisfying a threshold;
mark the die of the solid-state storage device as likely to fail, upon determining that the die of the solid-state storage device is likely to fail; and
transmit, from the solid-state storage device to the storage system controller, an indication that the die of the solid-state storage device has been marked as likely to fail.
2. The storage system of claim 1 , wherein the indication that the die of the solid-state storage device has been marked as likely to fail causes the storage system controller to place the die of the solid-state storage device in a read-only mode.
3. The storage system of claim 1 , wherein the indication that the die of the solid-state storage device has been marked as likely to fail causes the processing device to place the die of the solid-state storage device in a degraded mode.
4. The storage system of claim 1 , wherein the processing device is further to:
rebuild data stored at the die that is likely to fail, wherein the rebuilt data is to be transmitted to the storage system controller.
5. The storage system of claim 4 , wherein the processing device is further to:
transmit, to the storage system controller, a second indication that the rebuild operation on the data stored at the die was unsuccessful upon determining that the rebuild operation on the data stored at the die was unsuccessful.
6. The storage system of claim 1 , wherein the processing device is further to:
transmit data stored at the die that is likely to fail to the central storage system controller.
7. A method, comprising:
determining, by a processing device of a solid-state storage device of a plurality of solid-state storage devices each a comprising a respective processing device and a respective plurality of dies, whether a die that is one of a plurality of dies of the solid-state storage device of a storage system is likely to fail based on a number of failed input/output (I/O) operations performed on the die of the solid-state storage device;
marking, by the processing device of the solid-state storage device, the die of the solid-state storage device as likely to fail upon determining that the die is likely to fail; and
transmitting, from the solid-state storage device to a storage system controller of the storage system, an indication that the die of the solid-state storage device has been marked as likely to fail.
8. The method of claim 7 , wherein the indication that the die of the solid-state storage device has been marked as likely to fail causes the processing device to place the die of the solid-state storage device in a read-only mode.
9. The method of claim 7 , wherein the indication that the die of the solid-state storage device has been marked as likely to fail causes the storage system controller to place the die of the solid-state storage device in a degraded mode.
10. The method of claim 7 , further comprising:
rebuilding data stored at the die that is likely to fail, wherein the rebuilt data is to be transmitted to the storage system controller.
11. The method of claim 10 , further comprising:
transmitting, to the storage system controller, a second indication that the rebuild operation on the data stored at the die was unsuccessful upon determining that the rebuild operation on the data stored at the die was unsuccessful.
12. The method of claim 7 , further comprising:
transmitting data stored at the die that is likely to fail to the storage system controller.
13. A non-transitory computer readable storage medium storing instructions, which when executed, cause a processing device to:
determine whether a die that is one of a plurality of dies of a solid-state storage device of a storage system is likely to fail based on a number of errors associated with data stored at the die satisfies a threshold, wherein the solid-state storage device has the processing device as processing logic of the solid-state storage device;
mark the die of the solid-state storage device as likely to fail upon determining that the die of the solid-state storage device is likely to fail; and
transmit, from the solid-state storage device to a storage system controller of the storage system, an indication that the die of the solid-state storage device has been marked as likely to fail.
14. The non-transitory computer readable storage medium of claim 13 , wherein the indication that the die of the solid-state storage device has been marked as likely to fail causes the processing device to place the die of the solid-state storage device in a read-only mode.
15. The non-transitory computer readable storage medium of claim 13 , wherein the indication that the die of the solid-state storage device has been marked as likely to fail causes the storage system controller to place the die of the solid-state storage device in a degraded mode.
16. The non-transitory computer readable storage medium of claim 13 , wherein the processing device is further to:
rebuild the data stored at the die that is likely to fail, wherein the rebuilt data is to be transmitted to the storage system controller.
17. The non-transitory computer readable storage medium of claim 16 , wherein the processing device is further to:
transmit, to the storage system controller, a second indication that the rebuild operation on the data stored at the die was unsuccessful upon determining that the rebuild operation on the data stored at the die was unsuccessful.
18. The non-transitory computer readable storage medium of claim 13 , wherein the processing device is further to:
transmit the data stored at the die that is likely to fail to central storage system controller.