Method and system for improving flash storage utilization by predicting bad m-pages

• US 9,690,655 B2
• Filed: 09/30/2014
• Issued: 06/27/2017
• Est. Priority Date: 09/30/2014
• Status: Active Grant

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Inventors: Tabrizi, Haleh, Agarwal, Rajiv, Ferreira, Jeffrey Paul, Bonwick, Jeffrey S., Shapiro, Michael W.
• Primary Examiner(s): Riad, Amine
• Application Number:
  US14/501,917
• Publication Number:
  US 20160092304A1
• Time in Patent Office: 1,001 Days
• Field of Search: 714 61
• CPC Class Codes: G06F 11/0727 in a storage system, e.g. in a DASD or network based storage system drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/076 by exceeding a count or rate limit, e.g. word- or bit count limit
• CPC Class Codes: G06F 11/1072 in multilevel memories
• CPC Class Codes: G06F 11/108 Parity data distribution in semiconductor storages, e.g. in SSD
• CPC Class Codes: G06F 11/2221 to test input/output devices or peripheral units
• CPC Class Codes: G06F 2212/403 Error protection encoding, e.g. using parity or ECC codes
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/0679 Non-volatile semiconductor memory device, e.g. flash memory, one time programmable memory [OTP]
• CPC Class Codes: G06F 3/0688 Non-volatile semiconductor memory arrays

Abstract

A method for managing persistent storage. The method includes selecting a page for a proactive read request, where the page is located in the persistent storage. The method further includes issuing the proactive read request to the page, receiving, in response to the proactive read request, a bit error value (BEV) for data stored on the page, obtaining a BEV threshold (T) for the page, wherein T is determined using a program/erase cycle value associated with the page and a retention time of the data stored on the page, making a first determination that the BEV is greater than T, based on the first determination: identifying an m-page, where the m-page is a set of pages and the page is in the set of pages, and setting the m-page as non-allocatable for future operations.

Claims

1. A method for managing persistent storage, the method comprising:

selecting a page for a proactive read request, wherein the page is located in the persistent storage;

issuing the proactive read request to the page;

receiving, in response to the proactive read request, a bit error value (BEV) for data stored on the page;

obtaining a BEV threshold (T) for the page, wherein T is determined using a program/erase (P/E) cycle value associated with the page and a retention time of the data stored on the page, wherein obtaining T comprises performing a look-up in an in-memory data structure in memory, wherein the memory is located in a control module operatively connected to the persistent storage, wherein the in-memory data structure comprises a birth time for all data stored in the persistent storage and a plurality of entries, wherein each of the plurality of entries comprises one of a plurality of Ts, one of a plurality of P/E cycle values, and one of a plurality of retention times;

making a first determination that the BEV is greater than T;

based on the first determination;

identifying an m-page, wherein the m-page is a set of pages, wherein the page is in the set of pages;

setting the m-page as non-allocatable for future operations.

2. The method of claim 1, further comprising:

making a second determination that the data on the page is correctable using error-correcting codes;

based on the second determination;

scheduling the data on the page to be written to a new allocatable page in the persistent storage.

3. The method of claim 2, further comprising:

after the scheduling;

writing the data on the page to the new allocatable page as part of a garbage collection operation.

4. The method of claim 1, further comprising:

making a second determination that the data on the page is not correctable using error-correcting codes;

based on the second determination, making a third determination to proactively correct the data on the page using a RAID correction mechanism.

5. The method of claim 4, wherein the third determination takes into account a BEV of at least one other page in the persistent storage, wherein the page and the at least one other page are part of a RAID stripe.

6. The method of claim 1, wherein future operations comprise at least one selected from a group consisting of a write operation and a garbage collection operation.

7. The method of claim 1, wherein the BEV specifies a percentage of bits in the page that are incorrect.

8. The method of claim 1, wherein the BEV specifies a number of bits in the page that are incorrect.

9. The method of claim 1, wherein the BEV is determined using error-correcting codes.

10. The method of claim 1, wherein the page is located on a solid-state memory module in the persistent storage, wherein the solid-state memory module comprises a multi-level cell (MLC).

11. The method of claim 1, wherein the retention time is determined using a first time that the data was written to the page and a second time associated with the proactive read request.

12. The method of claim 1, wherein the P/E cycle value is a P/E cycle value range.

13. A system, comprising:

a storage module comprising a storage module controller and persistent storage; and

a control module operatively connected to the storage module and a client,

wherein control module comprises memory, wherein the memory comprises an in-memory data structure comprising a birth time for all data stored in the persistent storage and a plurality of entries, wherein each of the plurality of entries comprises one of a plurality of bit error value (BEV) thresholds (Ts), one of a plurality of P/E cycle values, and one of a plurality of retention times, and

wherein the retention time for the data stored on the page is determined using a birth time for the data stored on the page from the memory,

wherein the control module performs the following steps;

selecting a page for a proactive read request, wherein the page is located in the persistent storage;

issuing the proactive read request to the page;

obtaining, in response to the proactive read request, a BEV for data stored on the page;

obtaining a T for the page, wherein T is determined using a program/erase cycle value associated with the page and a retention time of the data stored on the page, wherein obtaining T comprises performing a look-up in the in-memory data structure;

making a first determination that the BEV is greater than T;

based on the first determination;

identifying an m-page, wherein the m-page is a set of pages, wherein the page is in the set of pages; and

setting the m-page as non-allocatable for future operations.

14. The system of claim 13, wherein the persistent storage comprises flash memory and wherein the page is located in the flash memory.

15. The system of claim 13, wherein the control module further performs the following steps:

making a second determination that the data on the page is correctable using error-correcting codes;

based on the second determination;

scheduling the data on the page to be written to a new allocatable page in the persistent storage;

after the scheduling, writing the data on the page to the new allocatable page as part of a garbage collection operation.

16. The system of claim 13, wherein the control module further performs the following steps:

making a second determination that the data on the page is not correctable using error-correcting codes;

based on the second determination, making a third determination to proactively correct the data on the page using a RAID correction mechanism.

17. The system of claim 16, wherein the third determination takes into account a BEV of at least one other page in the persistent storage, wherein the page and the at least one other page are part of a RAID stripe.

18. A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to:

select a page for a proactive read request, wherein the page is located in the persistent storage;

issue the proactive read request to the page;

receive, in response to the proactive read request, a bit error value (BEV) for data stored on the page;

obtain a BEV threshold (T) for the page, wherein T is determined using a program/erase (P/E) cycle value associated with the page and a retention time of the data stored on the page, wherein obtaining T comprises performing a look-up in an in-memory data structure in memory, wherein the memory is located in a control module operatively connected to the persistent storage, wherein the in-memory data structure comprises a birth time for all data stored in the persistent storage and a plurality of entries, wherein each of the plurality of entries comprises one of a plurality of Ts, one of a plurality of P/E cycle values, and one of a plurality of retention times;

make a first determination that the BEV is greater than T;

based on the first determination;

identify an m-page, wherein the m-page is a set of pages, wherein the page is in the set of pages; and

set the m-page as non-allocatable for future operations.