Systems and methods for adaptive proactive failure analysis for memories
In accordance with embodiments of the present disclosure, an information handling system may include a processor, a memory communicatively coupled to the processor and comprising a plurality of non-volatile memories, and a failure analysis module comprising a program of instructions, the failure analysis module configured to, when read and executed by the processor, set a predictive failure threshold for each of the plurality of non-volatile memories based at least on functional parameters of such non-volatile memory, and adapt the predictive failure threshold for each of the plurality of non-volatile memories based at least on health status parameters of such non-volatile memory.
1. An information handling system comprising:
a processor;
a memory communicatively coupled to the processor and comprising a plurality of non- volatile memories; and
a failure analysis module comprising a program of instructions, the failure analysis module configured to, when read and executed by the processor:
set a predictive failure threshold for each of the plurality of non-volatile memories based at least on functional parameters of each respective non-volatile memory, wherein the predictive failure threshold of each respective non-volatile memory is a leaky-bucket threshold for such non-volatile memory, such that when a number of correctable errors associated with such non-volatile memory exceeds the predictive failure threshold, the failure analysis module is configured to generate an indication that such non-volatile memory is susceptible to failure, wherein the functional parameters include operational features that are enabled or not enabled by each respective non-volatile memory, the operational features including at least one error-correcting code (ECC) feature, such that the predictive failure threshold is dependent on whether or not each operational feature is enabled by each respective non-volatile memory, wherein a first non-volatile memory for which the at least one ECC feature is enabled has a different predictive failure threshold than a second non-volatile memory for which the at least one ECC feature is not enabled; and
adapt the predictive failure threshold for each of the plurality of non-volatile memories based at least on health status parameters of each respective non-volatile memory.
2. The information handling system of claim 1 , the failure analysis module further configured to:
read, from each of the plurality of non-volatile memories, a health status register associated with each respective non-volatile memory and indicative of the health status parameters of each respective non-volatile memory; and
adapt the predictive failure threshold for each of the plurality of non-volatile memories based at least on information in the health status register of each respective non-volatile memory.
3. The information handling system of claim 1 , wherein each of the plurality of non-volatile memories comprises a storage class memory.
4. The information handling system of claim 1 , wherein the functional parameters of each respective non-volatile memory further comprise a memory type of each respective non-volatile memory.
5. The information handling system of claim 1 , wherein the health status parameters of each respective non-volatile memory comprise at least one of a number of spare sectors of each respective non-volatile memory remaining, an occurrence of thermal excursions within each respective non-volatile memory, and a number of correctable errors within media of each respective non-volatile memory itself.
6. A method comprising, in an information handling system comprising a processor and a memory communicatively coupled to the processor and comprising a plurality of non-volatile memories:
setting a predictive failure threshold for each of the plurality of non-volatile memories based at least on functional parameters of each respective non-volatile memory, wherein the predictive failure threshold of each respective non-volatile memory is a leaky-bucket threshold for such non-volatile memory, such that when a number of correctable errors associated with such non-volatile memory exceeds the predictive failure threshold, the failure analysis module is configured to generate an indication that such non-volatile memory is susceptible to failure, wherein the functional parameters include operational features that are enabled or not enabled by each respective non-volatile memory, the operational features including at least one error-correcting code (ECC) feature, such that the predictive failure threshold is dependent on whether or not each operational feature is enabled by each respective non-volatile memory, wherein a first non-volatile memory for which the at least one ECC feature is enabled has a different predictive failure threshold than a second non-volatile memory the at least one ECC feature is not enabled; and
adapting the predictive failure threshold for each of the plurality of non-volatile memories based at least on health status parameters of each respective non-volatile memory.
7. The method of claim 6 , further comprising:
reading, from each of the plurality of non-volatile memories, a health status register associated with each respective non-volatile memory and indicative of the health status parameters of each respective non-volatile memory; and
adapting the predictive failure threshold for each of the plurality of non-volatile memories based at least on information in the health status register of each respective non- volatile memory.
8. The method of claim 6 , wherein each of the plurality of non-volatile memories comprises a storage class memory.
9. The method of claim 6 , wherein the functional parameters of each respective non-volatile memory further comprise a memory type of each respective non-volatile memory.
10. The method of claim 6 , wherein the health status parameters of each respective non-volatile memory comprise at least one of a number of spare sectors of each respective non-volatile memory remaining, an occurrence of thermal excursions within each respective non-volatile memory, and a number of correctable errors within media of each respective non-volatile memory itself.
11. An article of manufacture comprising:
a non-transitory computer-readable medium; and
computer-executable instructions carried on the computer-readable medium, the instructions readable by a processor, the instructions, when read and executed, for causing the processor to, in an information handling system comprising a memory communicatively coupled to the processor and comprising a plurality of non-volatile memories:
set a predictive failure threshold for each of the plurality of non-volatile memories based at least on functional parameters of each respective non-volatile memory, wherein the predictive failure threshold of each respective non-volatile memory is a leaky-bucket threshold for such non-volatile memory, such that when a number of correctable errors associated with such non-volatile memory exceeds the predictive failure threshold, the failure analysis module is configured to generate an indication that such non-volatile memory is susceptible to failure, wherein the functional parameters include operational features that are enabled or not enabled by each respective non-volatile memory, the operational features including at least one error-correcting code (ECC) feature, such that the predictive failure threshold is dependent on whether or not each operational feature is enabled by each respective non-volatile memory, wherein a first non-volatile memory for which the at least one ECC feature is enabled has a different predictive failure threshold than a second non-volatile memory for which the at least one ECC feature is not enabled; and
adapt the predictive failure threshold for each of the plurality of non-volatile memories based at least on health status parameters of each respective non-volatile memory.
12. The article of claim 11 , the instructions for further causing the processor to:
read, from each of the plurality of non-volatile memories, a health status register associated with each respective non-volatile memory and indicative of the health status parameters of each respective non-volatile memory; and
adapt the predictive failure threshold for each of the plurality of non-volatile memories based at least on information in the health status register of each respective non-volatile memory.
13. The article of claim 11 , wherein each of the plurality of non-volatile memories comprises a storage class memory.
14. The article of claim 11 , wherein the functional parameters of each respective non-volatile memory further comprise a memory type of each respective non-volatile memory.
15. The article of claim 11 , wherein the health status parameters of each respective non-volatile memory comprise at least one of a number of spare sectors of each respective non-volatile memory remaining, an occurrence of thermal excursions within each respective non-volatile memory, and a number of correctable errors within media of each respective non-volatile memory itself.