Error characteristic estimation for NAND flash
Techniques related to improving a performance related to at least data reads from a memory are described. In an example, data is stored in a block of the memory as codewords. A data read includes a determination of whether each bit from a portion of the block is a zero or a one based on voltage measurements. Prior to decoding the codewords by performing a decoding procedure by an ECC decoder of the memory, a first number of errors “E 01 ” and a second number of errors “E 10 ” are estimated, where the first number of errors “E 01 ” is associated with bits each being a true zero and erroneously determined as a one, and where the second number of errors “E 10 ” associated with bits each being a true one and erroneously determined as a zero. Thereafter, the decoding of the codewords based on the decoding procedure is performed.
1. A method implemented on a computer system that includes a memory storing bits of a codeword and an error correction code (ECC) decoder decoding the codeword, the method comprising:
determining, based on voltage measurements of a hard decoding procedure, whether each bit stored in a portion of a block of the memory is a zero or a one;
prior to decoding the codeword by performing a soft decoding procedure by the ECC decoder, estimating a first number of errors “E 01 ” associated with bits each being a true zero and erroneously determined as a one and a second number of errors “E 10 ” associated with bits each being a true one and erroneously determined as a zero, the first number of errors “E 01 ” and the second number of errors “E 10 ” determined based on the voltage measurements of the hard decoding procedure, wherein estimating the first number of errors “E 01 ” comprises:
determining a first count “C 1 ” of the bits that are determined as a one based on the voltage measurements;
determining a second count “C 0 ” of the bits that are determined as a zero based on the voltage measurements; and
determining a flip bit count “E” as a function of a checksum of the bits that are determined based on the voltage measurement, the first number of errors “E 01 ” being based on the first count, the second count, and the flip bit count; and
performing the decoding of the codeword by the soft decoding procedure after the first number of errors “E 01 ” and the second number of errors “E 10 ” are estimated, the soft decoding procedure performed based on the first number of errors “E 01 ” and the second number of errors “E 10 ”.
2. The method of claim 1 , wherein the portion of the block is a page of the block.
3. The method of claim 2 , wherein the decoding of the first number of errors “E 01 ” and the second number of errors “E 10 ” are input to the soft decoding procedure.
4. The method of claim 1 , further comprising:
receiving a data write command to at least the portion of the block; and
updating a data write procedure to the portion of the block based on the first number of errors “E 01 ” and the second number of errors “E 10 ”.
5. The method of claim 1 , wherein determining whether each bit is a zero or a one is based on the hard decoding procedure that uses the voltage measurements.
6. The method of claim 1 , wherein estimating the first number of errors “E 01 ” and the second number of errors “E 10 ” comprises computing the checksum of the bits that are determined based on the voltage measurements.
7. The method of claim 1 , wherein estimating the first number of errors “E 01 ” and the second number of errors “E 10 ” comprises determining the first count of the bits that are determined as a one based on the voltage measurements and the second count of the bits that are determined as a zero based on the voltage measurements.
8. The method of claim 1 , wherein estimating the first number of errors “E 01 ” and the second number of errors “E 10 ” comprises determining the flip bit count.
9. The method of claim 1 , wherein determining whether each bit is a zero or a one is based on the hard decoding procedure that uses the voltage measurements, and wherein estimating the first number of errors “E 01 ” and the second number of errors “E 10 ” comprises:
computing the checksum of the bits that are determined based on the voltage measurements;
determining the first count “C 1 ” of the bits that are determined as a one based on the voltage measurements;
determining the second count “C 0 ” of the bits that are determined as a zero based on the voltage measurements; and
determining the flip bit count “E” as the function of the checksum, wherein the first number of errors “E 01 ” is based on the first count, the second count, and the flip bit count.
10. The method of claim 9 , wherein the function is a linear function.
11. The method of claim 10 , wherein the second number of errors “E 10 ” is estimated as (E−C 1 +C 0 )/2.
12. The method of claim 9 , wherein the first number of errors “E 01 ” is estimated as (C 1 −C 0 +E)/2.
13. A computer system comprising:
a memory storing bits of a codeword; and
an error correction code (ECC) decoder communicatively coupled with the memory and configured to:
determine, based on voltage measurements of a hard decoding procedure, whether each bit stored in a portion of a block of the memory is a zero or a one;
prior to decoding the codeword by performing a soft decoding procedure, estimate a first number of errors “E 01 ” associated with bits each being a true zero and erroneously determined as a one and a second number of errors “E 10 ” associated with bits each being a true one and erroneously determined as a zero, the first number of errors “E 01 ” and the second number of errors “E 10 ” determined based on the voltage measurements of the hard decoding wherein to estimate the first number of errors “E 01 ” and the second number of errors “E 10 ” comprises to:
compute a checksum of the bits that are determined based on the voltage measurements;
determine a first count “C 1 ” of the bits that are determined as a one based on the voltage measurements;
determine a second count “C 0 ” that are determined as a zero based on the voltage measurements; and
determine a flip bit count “E” as a function of the checksum, wherein the first number of errors “E 01 ” is based on the first count, the second count, and the flip bit count; and
perform the decoding of the codeword by the decoding procedure after the first number of errors “E 01 ” and the second number of errors “E 10 ” are estimated, the soft decoding procedure performed based on the first number of errors “E 01 ” and the second number of errors “E 10 ”.
14. The computer system of claim 13 , wherein the first number of errors “E 01 ” is estimated as (C 1 −C 0 +E)/2.
15. The computer system of claim 14 , wherein the second number of errors “E 10 ” is estimated as (E−C 1 +C 0 )/2.
16. A non-transitory computer-readable storage medium storing instructions that, upon execution on a computer system that includes a memory storing bits of a codeword and an error correction code (ECC) decoder decoding the codeword, cause the computer system to perform operations comprising:
determining, based on voltage measurements of a hard decoding procedure, whether each bit stored in a portion of a block of the memory is a zero or a one;
prior to decoding the codeword by performing a soft decoding procedure by the ECC decoder, estimating a first number of errors “E 01 ” associated with bits each being a true zero and erroneously determined as a one and a second number of errors “E 10 ” associated with bits each being a true one and erroneously determined as a zero, the first number of errors “E 01 ” and the second number of errors “E 10 ” determined based on the voltage measurements of the hard decoding procedure, wherein estimating the first number of errors “E 01 ” and the second number of errors “E 10 ” comprises:
computing a checksum of the bits that are determined based on the voltage measurements;
determining a first count “C 1 ” of the bits that are determined as a one based on the voltage measurements;
determining a second count “C 0 ” of the bits that are determined as a zero based on the voltage measurements; and
determining a flip bit count “E” as a function of the checksum, wherein the first number of errors “E 01 ” is based on the first count, the second count, and the flip bit count; and
performing the decoding of the codeword by the decoding procedure after the first number of errors “E 01 ” and the second number of errors “E 10 ” are estimated, the soft decoding procedure performed based on the first number of errors “E 01 ” and the second number of errors “E 10 ”.
17. The non-transitory computer-readable storage medium of claim 16 , wherein the first number of errors “E 01 ” is estimated as (C 1 −C 0 +E)/2, and wherein the second number of errors “E 10 ” is estimated as (E−C 1 +C 0 )/2.