Failed temperature sensor detection and mitigation within data storage devices
Methods and apparatus for detecting a failed temperature sensor within a data storage device and for mitigating the loss of the sensor are provided. One such data storage device includes a non-volatile memory (NVM), a set of temperature sensors, and a processor coupled to the NVM and the temperature sensors. The processor is configured to detect failure of one of the temperature sensors and obtain temperature data from the other temperature sensors. The processor is further configured to estimate, based on the obtained temperature data, the temperature at the failed sensor, and then control at least one function of the data storage device based on the estimated temperature, such as controlling entry into a Read Only mode. In some examples, the processor estimates the temperature at the failed sensor or at various virtual sensor locations using pre-determined formulas having offsets and coefficients determined during an initial machine learning calibration procedure.
1. A data storage device, comprising:
a non-volatile memory (NVM);
a plurality of temperature sensors; and
one or more processors coupled to the NVM and the plurality of temperature sensors, the one or more processors, individually or in combination, configured to:
detect a failure of one of the temperature sensors;
obtain temperature data from one or more of the other temperature sensors;
estimate, based on the obtained temperature data, the temperature at the failed temperature sensor;
estimate, based on the obtained temperature data and the estimated temperature at the failed temperature sensor, the temperature at a virtual temperature sensor location that is different from a location of the failed temperature sensor; and
control at least one function of the data storage device based on the estimated temperature at the virtual temperature sensor location.
2. The data storage device of claim 1 , wherein the one or more processors are further configured, individually or in combination, to:
determine whether the estimated temperature at the virtual temperature sensor location exceeds a threshold; and
responsive to the estimated temperature at the virtual temperature sensor location exceeding the threshold, control the data storage device to enter a Read Only mode.
3. The data storage device of claim 1 ,
wherein the plurality of temperature sensors includes an NVM temperature sensor configured to measure the temperature of the NVM, and
wherein the one or more processors are further configured, individually or in combination, to estimate, in response to a failure of the NVM temperature sensor, the temperature of the NVM from the temperature data obtained from the other temperature sensors and the temperature at the virtual temperature sensor location.
4. The data storage device of claim 3 , wherein the one or more processors are further configured, individually or in combination, to execute write commands to write data in the NVM despite the failure of the NVM temperature sensor so long as the estimated temperature of the NVM does not exceed a threshold that triggers a Read Only mode.
5. The data storage device of claim 1 , wherein the one or more processors are further configured, individually or in combination, to:
store a formula in a memory of the data storage device for estimating the temperature at the failed temperature sensor, the formula derived from historical temperature data from the one or more of the other temperature sensors; and
estimate the temperature at the failed temperature sensor using the formula.
6. The data storage device of claim 5 , wherein the one or more processors are further configured, individually or in combination, to:
determine the formula for estimating the temperature at the failed temperature sensor based on the historical temperature data using a machine learning model.
7. The data storage device of claim 1 , wherein the one or more processors are further configured, individually or in combination, to:
store a formula in a memory of the data storage device for estimating the temperature at the virtual temperature sensor location; and
estimate the temperature at the virtual temperature sensor location using the formula.
8. The data storage device of claim 1 , wherein the one or more processors are further configured, individually or in combination, to:
estimate the temperature at the failed temperature sensor based on temperature data obtained from a nearest of the one or more other temperature sensors to the failed temperature sensor.
9. The data storage device of claim 1 , wherein the one or more processors are further configured, individually or in combination, to:
estimate the temperature at the failed temperature sensor based on an average of the temperature data obtained from the one or more other temperature sensors.
10. The data storage device of claim 1 , wherein the plurality of temperature sensors comprises:
a first sensor configured to measure the temperature of the NVM;
a second sensor configured to measure the temperature of the one or more processors; and
a third sensor configured to measure the temperature of a clock generator of the data storage device.
11. The data storage device of claim 1 , wherein the one or more processors are further configured, individually or in combination, to control the at least one function of the data storage device based on the estimated temperature at the virtual temperature sensor location and the estimated temperature at the failed temperature sensor, while using different triggering-thresholds for the estimated temperature at the virtual temperature sensor location and the estimated temperature at the failed temperature sensor.
12. A method for use by a data storage device having a non-volatile memory (NVM) and a plurality of temperature sensors, the method comprising:
detecting a failure of one of the temperature sensors;
obtaining temperature data from one or more of the other temperature sensors;
estimating, based on the obtained temperature data, the temperature at the failed temperature sensor;
estimating, based on the obtained temperature data and the estimated temperature at the failed temperature sensor, the temperature at a virtual temperature sensor location that is different from a location of the failed temperature sensor; and
controlling at least one function of the data storage device based on the estimated temperature at the virtual temperature sensor location.
13. The method of claim 12 , further comprising:
determining whether the estimated temperature at the virtual temperature sensor location exceeds a threshold; and
responsive to the estimated temperature at the virtual temperature sensor location exceeding the threshold, controlling the data storage device to enter a Read Only mode.
14. The method of claim 12 , wherein the plurality of temperature sensors includes an NVM temperature sensor configured to measure the temperature of the NVM, and wherein the method further comprises estimating, in response to a failure of the NVM temperature sensor, the temperature of the NVM from the temperature data obtained from the other temperature sensors and the temperature at the virtual temperature sensor location.
15. The method of claim 14 , further comprising executing write commands to write data in the NVM despite the failure of the NVM temperature sensor so long as the estimated temperature of the NVM does not exceed a threshold that triggers a Read Only mode.
16. The method of claim 12 , further comprising:
storing a formula within a memory of the data storage device for estimating the temperature at the failed temperature sensor, the formula derived from historical temperature data from the one or more of the other temperature sensors; and
estimating the temperature at the failed temperature sensor using the formula.
17. The method of claim 16 , further comprising:
determining the formula for estimating the temperature at the failed temperature sensor based on the historical temperature data using a machine learning model.
18. The method of claim 12 , further comprising:
storing a formula within a memory of the data storage device for estimating the temperature at a virtual temperature sensor location; and
estimating the temperature at the virtual temperature sensor location using the formula.
19. The method of claim 12 , wherein controlling the at least one function comprises controlling the at least one function based on the estimated temperature at the virtual temperature sensor location and the estimated temperature at the failed temperature sensor, while using different triggering-thresholds for the estimated temperature at the virtual temperature sensor location and the estimated temperature at the failed temperature sensor.
20. An apparatus for use with a data storage device having a non-volatile memory (NVM) and a plurality of temperature sensors, the apparatus comprising:
means for detecting a failure of one of the temperature sensors;
means for obtaining temperature data from one or more of the other temperature sensors;
means for estimating, based on the obtained temperature data, the temperature at the failed temperature sensor;
means for estimating, based on the obtained temperature data and the estimated temperature at the failed temperature sensor, the temperature at a virtual temperature sensor location that is different from a location of the failed temperature sensor; and
means for controlling at least one function of the data storage device based on the estimated temperature at the virtual temperature sensor location.