Storage system and method for data-driven intelligent thermal throttling
A storage system and method for data-driven intelligent thermal throttling are provided. In one embodiment, the storage system comprises a memory and a controller. The controller is configured to determine a temperature of the memory, estimate a future temperature curve based on the temperature of the memory, and determine a memory throttling delay to apply based on the estimated future temperature curve. Other embodiments are provided.
1. A storage system comprising:
a memory; and
a controller configured to:
take samples of temperatures of the memory over a rolling window;
dynamically change, based on a temperature of the memory, a length of the rolling window, a frequency of temperature sampling of the memory in the rolling window, and a movement of the rolling window over time;
generate estimated future temperature curves from the samples of temperatures of the memory;
determine memory throttling delays to apply based on each estimated future temperature curve; and
reduce a number of program command sequences sent to the memory and a number of program operations performed by the memory by dynamically applying the memory throttling delays based on each estimated future temperature curve, wherein each estimated future temperature curve is associated with a relationship between performance and saturation temperature, which is used to determine the memory throttling delays.
2. The storage system of claim 1 , wherein the controller is further configured to estimate each future temperature curve based on the saturation temperature and a curvature factor.
3. The storage system of claim 1 , wherein the controller is further configured to estimate each future temperature curve by minimizing a sum of squares of errors.
4. The storage system of claim 1 , wherein the controller is further configured to estimate a new future temperature curve after applying the memory throttling delays.
5. The storage system of claim 1 , wherein the controller is further configured to apply static throttling as an exception handling operation.
6. The storage system of claim 1 , wherein the memory comprises a three-dimensional memory.
7. The storage system of claim 1 , wherein the storage system is configured to be embedded in a host.
8. The storage system of claim 1 , wherein the storage system is configured to be removably connected to a host.
9. The storage system of claim 1 , wherein the controller is further configured to dynamically apply the memory throttling delays based on each estimated future temperature curve rather than applying memory throttling delays based on fixed, pre-defined thresholds.
10. A method comprising:
performing the following in a storage system:
taking samples of temperatures of a memory over a rolling window;
dynamically changing, based on a temperature of the memory, a length of the rolling window, a frequency of temperature sampling of the memory in the rolling window, and a movement of the rolling window over time;
generating estimated future temperature curves from the samples of temperatures of the memory;
determining memory throttling delays to apply based on each estimated future temperature curve; and
reducing a number of program command sequences sent to the memory and a number of program operations performed by the memory by dynamically applying the memory throttling delays based on each estimated future temperature curve, wherein each estimated future temperature curve is associated with a relationship between performance and saturation temperature, which is used to determine the memory throttling delays.
11. The method of claim 10 , wherein each future temperature curve is estimated based on an actual temperature of the memory, the saturation temperature, and a curvature factor.
12. The method of claim 10 , further comprising using a minimizing-a-sum-of-squares-of-errors technique to estimate each future temperature curve.
13. The method of claim 10 , further comprising estimating a new future temperature curve after applying the memory throttling delays.
14. The method of claim 10 , further comprising applying static throttling in response to an exception.
15. The method of claim 10 , wherein the storage system is configured to be embedded in a host.
16. The method of claim 10 , wherein the memory throttling delays are dynamically applied based on each estimated future temperature curve rather than applying memory throttling delays based on fixed, pre-defined thresholds.
17. A storage system comprising:
a memory;
means for taking samples of temperatures of the memory over a rolling window;
means for generating estimated future temperature curves from the samples of temperatures of the memory;
means for determining memory throttling delays to apply based on each estimated future temperature curve; and
means for reducing a number of program command sequences sent to the memory and a number of program operations performed by the memory by dynamically applying the memory throttling delays based on each estimated future temperature curve, wherein each estimated future temperature curve is associated with a relationship between performance and saturation temperature, which is used to determine the memory throttling delays.