Proactive corrective actions in memory based on a probabilistic data structure
The present disclosure includes apparatuses and methods for proactive corrective actions in memory based on a probabilistic data structure. A number of embodiments include a memory, and circuitry configured to input information associated with a subset of data stored in the memory into a probabilistic data structure and proactively determine, at least partially using the probabilistic data structure, whether to take a corrective action on the subset of data stored in the memory.
1. An apparatus, comprising:
a memory; and
circuitry configured to:
input identification information associated with a subset of data stored in the memory into a probabilistic data structure;
perform, using the probabilistic data structure, an operation on the inputted identification information, wherein the operation includes:
processing the inputted identification information by a number of hash functions of the probabilistic data structure, wherein an output of each respective hash function corresponds to a different counter of the probabilistic data structure; and
updating each of the different counters to which the output of each respective hash function corresponds by a quantity of pages in the subset of data on which a sense operation was performed; and
proactively determine whether to take a corrective action on the subset of data based, at least in part, on a result of the operation performed by the probabilistic data structure, wherein the result of the operation performed by the probabilistic data structure corresponds to a value of each of the different counters after they have been updated by the quantity of pages.
2. The apparatus of claim 1 , wherein the operation includes updating each of the different counters by a quantity of times the subset of data was accessed during a sense or program operation performed on at least a portion of the subset of data.
3. The apparatus of claim 1 , wherein the operation includes updating each of the different counters by a quantity of program and/or erase cycles performed on at least a portion of the subset of data.
4. The apparatus of 1 wherein the circuitry is configured to proactively determine whether to take the corrective action based on whether the value of each of the different counters after they have been updated meets or exceeds a particular threshold.
5. The apparatus of claim 1 , wherein updating each of the different counters includes increasing the value of each of the different counters.
6. The apparatus of claim 1 , wherein updating each of the different counters includes decreasing the value of each of the different counters.
7. The apparatus of claim 1 , wherein updating each of the different counters includes resetting a value of each of the different counters.
8. A method for operating memory, comprising:
inputting identification information associated with a subset of data stored in a memory into a counting Bloom filter having a number of counters and a number of hash functions;
performing, by the counting Bloom filter, an operation on the inputted identification information, wherein the operation includes:
processing the inputted identification information by the number of hash functions, wherein an output of each respective hash function corresponds to a different one of the number of counters; and
updating each of the different counters to which the output of each respective hash function corresponds by a quantity of pages in the subset of data on which a sense operation was performed; and
proactively determining whether to take a corrective action on the subset of data based, at least in part, on a result of the operation performed by the counting Bloom filter, wherein the result of the operation performed by the counting Bloom filter corresponds to a value of each of the different counters after they have been updated by the quantity of pages.
9. The method of claim 8 , wherein the method includes inputting the identification information into the counting Bloom filter upon determining the identification information is not stored in a cache.
10. The method of claim 8 , wherein:
the identification information is inputted into the counting Bloom filter in response to the sense operation being performed on the subset of data.
11. The method of claim 8 , wherein:
the identification information is inputted into the counting Bloom filter in response to the sense operation being performed on the subset of data; and
the operation includes updating each of the different counters by a quantity of times the subset of data is accessed during the sense operation.
12. The method of claim 8 , wherein:
the identification information is inputted into the counting Bloom filter in response to a program operation being performed on the subset of data; and
the operation includes updating each of the different counters by a quantity of times the subset of data is accessed during the program operation.
13. The method of claim 8 , wherein the number of hash functions is two.
14. An apparatus, comprising:
a memory; and
circuitry configured to:
perform a program operation or a sense operation on a subset of data stored in the memory;
input identification information associated with the subset of data into a probabilistic data structure;
perform, using the probabilistic data structure, an operation on the inputted identification information, wherein the operation includes:
processing the inputted identification information by a number of hash functions of the probabilistic data structure, wherein an output of each respective hash function corresponds to a different counter of the probabilistic data structure;
updating each of the different counters to which the output of each respective hash function corresponds by a quantity of pages in the subset of data on which a sense operation was performed if a sense operation was performed on the subset of data; and
resetting each of the different counters to which the output of each respective hash function corresponds if a program operation was performed on the subset of data; and
proactively determine whether to take a corrective action on the subset of data based, at least in part, on a result of the operation performed using the probabilistic data structure, wherein the result of the operation performed by the probabilistic data structure corresponds to a value of each of the different counters after they have been updated by the quantity of pages.
15. The apparatus of claim 14 , wherein proactively determining whether to take the corrective action includes determining whether to take the corrective action before a failure of the memory occurs.
16. The apparatus of claim 14 , wherein the probabilistic data structure is a counting Bloom filter.
17. The apparatus of claim 14 , wherein the subset of data is a block of data.
18. The apparatus of claim 14 , wherein resetting each of the different counters includes setting a value of each of the different counters to zero.