Predictive performance indicator for storage devices
Systems and methods for predictive performance indicators for storage devices are described. The data storage device may process host storage operations and maintenance operations that impact real-time performance. A performance value and corresponding threshold may be determined. Increases in maintenance operations and resulting changes in the performance value may be predicted. When the predicted change in performance value crosses the performance threshold, the host device may be notified.
1. A system comprising:
a data storage device comprising:
a storage medium configured to store host data; and
a storage device controller comprising processing resources and configured to:
process, using the storage medium and the processing resources, storage operations from a host device;
process, using the storage medium and the processing resources, maintenance operations for increasing storage availability in the storage medium;
determine a performance value for the storage operations;
predict an increase in maintenance operations;
predict, based on the predicted increase in maintenance operations, a change in the performance value; and
notify, through a host interface, the host device of the predicted change in the performance value.
2. The system of claim 1 , wherein:
the performance value is selected from:
an input/output operations per unit time value;
a data volume per unit time value; and
a nominal performance value for the data storage device;
the storage device controller is further configured to:
determine a performance threshold for the performance value; and
determine that the change in the performance value crosses the performance threshold; and
notifying the host device of the predicted change in the performance value is responsive to determining that the change in the performance value crosses the performance threshold.
3. The system of claim 1 , wherein the storage device controller is further configured to:
receive, through the host interface, a host performance window value;
determine, based on the host performance window value, a performance window; and
predict the increase in maintenance operations over the performance window.
4. The system of claim 1 , wherein the storage device controller is further configured to:
receive, through the host interface, a host prediction horizon value;
determine, based on the host prediction horizon value, a prediction horizon; and
predict the increase in maintenance operations at a future time corresponding to the prediction horizon.
5. The system of claim 1 , wherein the storage device controller is further configured to:
determine a plurality of severity thresholds;
determine, based on the predicted change in the performance value and the plurality of severity thresholds, a severity value for the predicted change; and
report, through the host interface, the severity value for the predicted change.
6. The system of claim 1 , wherein the storage device controller is further configured to:
determine a prediction probability value for the predicted change in the performance value; and
report, through the host interface, the prediction probability value for the predicted change.
7. The system of claim 1 , wherein:
the host interface includes a notification register; and
the host device is configured to:
periodically check the notification register; and
receive, from the data storage device and responsive to checking the notification register, notification of the predicted change in the performance value.
8. The system of claim 1 , wherein:
the host interface includes an interface function register comprising a plurality of memory locations; and
the storage device controller is further configured to store, responsive to the predicted change in the performance value, at least one performance change indicator value in an indicator memory location of the plurality of memory locations in the interface function register.
9. The system of claim 1 , further comprising the host device, wherein:
the host device comprises:
a host processor;
a host memory:
a storage interface configured to communicate with the host interface of the data storage device; and
a storage manager configured to:
send storage operations to the data storage device according to a defined host access pattern and a storage performance requirement;
receive, from the data storage device, notification of the predicted change in the performance value; and
display, to a user of the host device, a warning notification indicating the predicted change in the performance value.
10. The system of claim 9 , wherein:
the data storage device comprises a removable storage device configured to be removably attached to the host device;
the host device is configured to initialize the data storage device;
the storage manager is further configured to:
establish, responsive to initializing the data storage device, a performance interface with the data storage device; and
set, through the performance interface, at least one performance monitoring parameter selected from:
a performance value type;
a host performance threshold value;
a host performance window value;
a host prediction horizon value;
a notification type; and
a reporting interval;
the storage device controller is further configured to:
determine, responsive to the data storage device being initialized and based on a plurality of storage operations from the host, the defined host access pattern; and
predict the increase in maintenance operations based on the defined host access pattern continuing through a performance window; and
the maintenance operations for increasing storage availability include garbage collection of flash memory data blocks.
11. A computer-implemented method, comprising:
processing, by a data storage device, storage operations from a host device;
processing, by the data storage device, maintenance operations for increasing storage availability in a storage medium in the data storage device;
determining, by the data storage device, a performance value for the storage operations;
predicting, by the data storage device, an increase in maintenance operations;
predicting, by the data storage device and based on the predicted increase in maintenance operations, a change in the performance value; and
notifying, by the data storage device and through a host interface, the host device of the predicted change in the performance value.
12. The computer-implemented method of claim 11 , further comprising:
determining, by the data storage device, a performance threshold for the performance value; and
determining, by the data storage device, that the change in the performance value crosses the performance threshold, wherein notifying the host device of the predicted change in the performance value is responsive to determining that the change in the performance value crosses the performance threshold.
13. The computer-implemented method of claim 11 , further comprising:
receiving, by the data storage device and through the host interface, a host performance window value;
determining, by the data storage device and based on the host performance window value, a performance window; and
predicting, by the data storage device, the increase in maintenance operations over the performance window.
14. The computer-implemented method of claim 11 , further comprising:
receiving, by the data storage device and through the host interface, a host prediction horizon value;
determining, by the data storage device and based on the host prediction horizon value, a prediction horizon; and
predicting, by the data storage device, the increase in maintenance operations at a future time corresponding to the prediction horizon.
15. The computer-implemented method of claim 11 , further comprising:
determining, by the data storage device, a plurality of severity thresholds;
determining, by the data storage device and based on the predicted change in the performance value and the plurality of severity thresholds, a severity value for the predicted change; and
reporting, by the data storage device and through the host interface, the severity value for the predicted change.
16. The computer-implemented method of claim 11 , further comprising:
determining, by the data storage device, a prediction probability value for the predicted change in the performance value; and
reporting, by the data storage device and through the host interface, the prediction probability value for the predicted change.
17. The computer-implemented method of claim 11 , further comprising:
storing, by the data storage device, a notification value in a notification register of the host interface;
periodically checking, by the host device, the notification register; and
receiving, by the host device and responsive to checking the notification register, notification of the predicted change in the performance value.
18. The computer-implemented method of claim 11 , further comprising:
storing, by the data storage device and responsive to the predicted change in the performance value, at least one performance change indicator value in an indicator memory location, wherein the indicator memory location is in an interface function register in the host interface.
19. The computer-implemented method of claim 11 , further comprising:
sending, by the host device, storage operations to the data storage device according to a defined host access pattern and a storage performance requirement;
receiving, by the host device and from the data storage device, notification of the predicted change in the performance value; and
displaying, by the host device and to a user of the host device, a warning notification indicating the predicted change in the performance value.
20. A data storage device, comprising:
a storage medium configured to store host data;
a storage device controller comprising:
a processor;
an operating memory; and
a host interface configured for communication with a host device;
means for processing storage operations from the host device;
means for processing maintenance operations for increasing storage availability in a storage medium in the data storage device;
means for determining a performance value for the storage operations;
means for determining a performance threshold for the performance value;
means for predicting an increase in maintenance operations;
means for predicting, based on the predicted increase in maintenance operations, a change in the performance value that crosses the performance threshold; and
means for notifying, through the host interface, the host device of the predicted change in the performance value.