Rapid restart protection for a non-volatile memory system

Patent metadata

Field	Value
Publication number	US-11094381-B2
Application number	US-201916429021-A
Country	US
Kind code	B2
Filing date	Jun 2, 2019
Priority date	Jun 2, 2019
Publication date	Aug 17, 2021
Grant date	Aug 17, 2021

Abstract

Systems and methods for managing non-volatile memory devices are provided. Embodiments discussed herein provide rapid restart protection for journaling system. The rapid restart protection prevents the NVM from experiencing memory saturation when the NVM system is being forced to handle multiple successive restarts.

First claim

What is claimed is:

1. A method for protecting a non-volatile memory (NVM) of a computing device during a rapid restart scenario, the method comprising, at the computing device:
determining that a current journal of the NVM includes a rapid restart key, wherein the rapid restart key indicates that a prior version of indirection data stored on the NVM has not been updated to a current version of indirection data;
deleting at least one band that includes at least the current journal and the rapid restart key;
establishing, within at least one new band of the NVM, the current journal and the rapid restart key;
attempting to establish, within the at least one new band, the current version of the indirection data based on a prior journal of the NVM and the prior version of indirection data; and
in response to determining that the current version of the indirection data is established within the at least one new band:
deleting one or more bands of the NVM that include the prior version of indirection data and the prior journal.

2. The method of claim 1 , wherein the at least one band further includes an incomplete copy of the current version of indirection data.

3. The method of claim 1 , wherein the at least one band and the at least one new band are:
distinct from one another, or
the same as one another.

4. The method of claim 1 , further comprising:
prior to determining that the current journal of the NVM includes the rapid restart key:
disabling host write transactions; and

subsequent to deleting the one or more bands of the NVM:
enabling host write transactions.

5. The method of claim 1 , further comprising, in response to determining that the current version of the indirection data is established within the at least one new band:
writing a flag into current journal to indicate that the prior version of the indirection data has successfully been updated to the current version of the indirection data.

6. The method of claim 1 , wherein the indirection data comprises a logical-to-physical mapping of data stored on the NVM.

7. The method of claim 1 , wherein the rapid restart scenario is caused by a sudden power failure.

8. The method of claim 1 , wherein the prior and current versions of indirection data are separated into a first plurality of silos and a second plurality of silos, respectively.

9. A computing device configured to protect a non-volatile memory (NVM) during a rapid restart scenario, the computing device comprising:
at least one processor; and
at least one memory storing instructions that, when executed by the at least one processor, cause the computing device to carry out steps that include:
determining that a current journal of the NVM includes a rapid restart key, wherein the rapid restart key indicates that a prior version of indirection data stored on the NVM has not been updated to a current version of indirection data;

deleting at least one band that includes at least the current journal and the rapid restart key;
establishing, within at least one new band of the NVM, the current journal and the rapid restart key;
attempting to establish, within the at least one new band, the current version of the indirection data based on a prior journal of the NVM and the prior version of indirection data; and
in response to determining that the current version of the indirection data is established within the at least one new band:
deleting one or more bands of the NVM the include the prior version of indirection data and the prior journal.

10. The computing device of claim 9 , wherein the at least one band further includes an incomplete copy of the current version of indirection data.

11. The computing device of claim 9 , wherein the at least one band and the at least one new band are:
distinct from one another, or
the same as one another.

12. The computing device of claim 9 , wherein the steps further include:
prior to determining that the current journal of the NVM includes the rapid restart key:
disabling host write transactions; and

subsequent to deleting the one or more bands of the NVM:
enabling host write transactions.

13. The computing device of claim 9 , wherein the steps further include, in response to determining that the current version of the indirection data is established within the at least one new band:
writing a flag into current journal to indicate that the prior version of the indirection data has successfully been updated to the current version of the indirection data.

14. The computing device of claim 9 , wherein the indirection data comprises a logical-to-physical mapping of data stored on the NVM.

15. The computing device of claim 9 , wherein the rapid restart scenario is caused by a sudden power failure.

16. The computing device of claim 9 , wherein the prior and current versions of indirection data are separated into a first plurality of silos and a second plurality of silos, respectively.

17. A method for protecting a non-volatile memory (NVM) of a computing device during a rapid restart scenario, the method comprising, at the computing device:
resetting a rapid restart counter; and
in response to a restart event:
incrementing the rapid restart counter,
commencing a failure protocol when the rapid restart counter satisfies a first threshold,
starting a rapid restart protection process when the rapid restart counter satisfies neither the first threshold nor a second threshold, and
the rapid restart protection protocol process prevents bricking of the NVM, and

delaying the rapid restart protection process for a fixed period of time when the rapid restart counter does not satisfy the first threshold but satisfies the second threshold, wherein the failure protocol prevents operation of the rapid restart protection process.

18. The method of claim 17 , wherein the rapid restart protection process erases all NVM bands written to during an immediately preceding restart process if the immediately preceding restart process was unsuccessful.

19. The method of claim 17 , wherein the rapid restart counter is reset in response to a successful restart.

20. The method claim 19 , wherein:
the fixed period of time enables a cause of a sudden power failure to resolve, the sudden power failure is responsible for the restart, and
the first threshold is greater than the second threshold.

Assignees

• Apple Inc

Inventors

• Ashraf Muhammad N
• Paley Alexander
• Liu Yuhua
• Khmelnitsky Vadim
• Byom Matthew J

Classifications

• G11C16/22 Primary

  Safety or protection circuits preventing unauthorised or accidental access to memory cells · CPC title

• G11C7/109

  Control signal input circuits · CPC title

• G06F1/28

  Supervision thereof, e.g. detecting power-supply failure by out of limits supervision · CPC title

• G06F1/30 Primary

  Means for acting in the event of power-supply failure or interruption, e.g. power-supply fluctuations (for resetting only G06F1/24) · CPC title

• G11C16/16

  for erasing blocks, e.g. arrays, words, groups · CPC title