Bad block detection and predictive analytics in NAND flash storage devices

• US 10,146,604 B2
• Filed: 08/23/2016
• Issued: 12/04/2018
• Est. Priority Date: 08/23/2016
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Rudy, David, Kechriotis, George, O'Grady, Patrick, Gemmell, James
• Primary Examiner(s): Bonzo, Bryce P
• Assistant Examiner(s): Arcos, Jeison C
• Application Number:
  US15/244,190
• Publication Number:
  US 20180060148A1
• Time in Patent Office: 833 Days
• 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 11/0769 Readable error formats, e.g. cross-platform generic formats, human understandable formats
• CPC Class Codes: G06F 11/0772 Means for error signaling, e.g. using interrupts, exception flags, dedicated error registers
• CPC Class Codes: G06F 11/0787 Storage of error reports, e.g. persistent data storage, storage using memory protection
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/24 Marginal checking or other specified testing methods not covered by G06F11/26, e.g. race tests
• CPC Class Codes: G06F 2201/805 Real-time
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G11C 29/38 Response verification devices
• CPC Class Codes: G11C 29/44 Indication or identification of errors, e.g. for repair
• CPC Class Codes: G11C 29/52 Protection of memory contents; Detection of errors in memory contents

Abstract

Utilities for use in actively detecting the occurrence of bad blocks in NAND flash storage devices and diagnosing the devices as faulty at some point before complete failure of the devices (e.g., before a number of allowable bad blocks has been reached) to allow a corresponding service processor to continue to write to available blocks for a period of time until a replacement NAND flash device can be identified. The utilities may also be utilized to predict the future occurrence of bad blocks in NAND flash devices, such as during the “burn-in” process of the devices (e.g., which tests the quality of the NAND flash device before being placed into service to weed out devices with defects).

Claims

1. A method for use with a non-volatile storage device, comprising;

obtaining a count of allowable bad blocks for an array of a non-volatile storage device (“

allowable bad block count”

), wherein the allowable bad block count equals a count of blocks in the non-volatile storage device set aside to receive data intended for a block marked as being a bad block;

obtaining a count of blocks in the array that have been marked as being bad blocks (“

current bad block count”

), wherein each block marked as being a bad block contains one or more invalid bits;

determining, with a processor, a percentage of the allowable bad block count represented by the current bad block count (“

allowable bad block percentage”

);

ascertaining, with the processor, whether the allowable bad block percentage is greater than a threshold percentage;

generating a message based on a result of the ascertaining step;

conducting a burn-in test of the array of the non-volatile storage device, wherein the steps of obtaining the current bad block count, determining the allowable bad block percentage, ascertaining, and generating are performed during the conducting step, and wherein the conducting step includes;

first conducting a first phase of the burn-in test for a first length of time, wherein the threshold percentage is a first threshold percentage during the first conducting step, and wherein the first threshold percentage is selected as a function of the first length of time; and

second conducting a second phase of the burn-in test for a second length of time, wherein the threshold percentage is a second threshold percentage during the second conducting step, wherein the second threshold percentage is selected as a function of the second length of time, and wherein the first and second lengths of time are different.

2. The method of claim 1, wherein the generating step includes generating a message when the processor ascertains that the allowable bad block percentage is greater than the threshold percentage.

3. The method of claim 2, wherein the message indicates diagnosis of a fault in the non-volatile storage device.

4. The method of claim 2, further including:

initiating a failover process for the processor.

5. The method of claim 1, further including:

obtaining a total count of blocks in the array that have been marked as being bad blocks (“

total current bad block count”

);

obtaining a total count of allowable bad blocks for the array (“

total allowable bad block count”

); and

obtaining a count of blocks in the array that were marked as being bad blocks upon an initial boot of an operating system of the processor (“

initial bad block count”

), wherein the current bad block count is the difference between the total current bad block count and the initial bad block count, and wherein the allowable bad block count is the difference between the total allowable bad block count and the initial bad block count.

6. The method of claim 1, wherein the generating step includes generating a message when the processor ascertains that the allowable bad block percentage is greater than the threshold percentage, and wherein the message indicates diagnosis of a fault in the non-volatile storage device.

7. The method of claim 1, wherein:

during the first conducting step, the ascertaining step includes ascertaining whether the allowable bad block percentage is greater than an upper first threshold percentage and/or a lower first threshold percentage, wherein the generating step includes generating a message without an indication of a fault when the allowable bad block percentage is greater than the upper first threshold percentage but not greater than the lower first threshold percentage, and wherein the generating step includes generating a message with an indication of a fault when the allowable bad block percentage is greater than the upper first threshold percentage and the lower first threshold percentage; and

during the second conducting step, the ascertaining step includes ascertaining whether the allowable bad block percentage is greater than an upper second threshold percentage and/or a lower second threshold percentage, wherein the generating step includes generating a message without an indication of a fault when the allowable bad block percentage is greater than the lower second threshold percentage but not greater than the upper second threshold percentage, and wherein the generating step includes generating a message with an indication of a fault when the allowable bad block percentage is greater than the lower second threshold percentage and the upper second threshold percentage.

8. A method, comprising:

first conducting a first phase of a burn-in test on a NAND flash storage device that includes a storage array with a series of blocks for storing data;

first determining, during the first conducting step, whether a current percentage of a count of allowable bad blocks for the series of blocks (“

allowable bad block count”

) represented by a current count of bad blocks in the series of blocks (“

current bad block count”

) has reached or exceeded a first threshold percentage of the allowable bad block count, wherein each bad block contains one or more invalid bits, and wherein the allowable bad block count equals a count of blocks in the series of blocks that are set aside to receive data intended for a bad block;

second conducting, after the first conducting step, a second phase of a burn-in test on the NAND flash storage device;

second determining, during the second conducting step, whether the current percentage has reached or exceeded a second threshold percentage of the allowable bad block count; and

generating at least one message when the current percentage reaches or exceeds the first threshold percentage during the first conducting step or the second threshold percentage during the second conducting step.

9. The method of claim 8, wherein a length of the first phase is different than a length of the second phase.

10. The method of claim 8, wherein the first and second threshold percentages are different.

11. The method of claim 8, further including:

obtaining a total count of blocks in the storage array that have been marked as being bad blocks (“

total current bad block count”

);

obtaining a total count of allowable bad blocks for the storage array (“

total allowable bad block count”

); and

obtaining a count of blocks in the storage array that were marked as being bad blocks upon an initial boot of an operating system of the NAND flash storage device, wherein the current bad block count is the difference between the total current bad block count and the initial bad block count, and wherein the allowable bad block count is the difference between the total allowable bad block count and the initial bad block count.

12. The method of claim 8, wherein the first determining includes first determining whether the current percentage is greater than an upper first threshold percentage and/or a lower first threshold percentage, wherein the generating step includes generating a message without an indication of a fault when the current percentage is greater than the upper first threshold percentage but not greater than the lower first threshold percentage, and wherein the generating step includes generating a message with an indication of a fault when the current percentage is greater than the upper first threshold percentage and the lower first threshold percentage.

13. The method of claim 12, wherein the second determining includes second determining whether the current percentage is greater than an upper second threshold percentage and/or a lower second threshold percentage, wherein the generating step includes generating a message without an indication of a fault when the current percentage is greater than the upper second threshold percentage but not greater than the lower second threshold percentage, and wherein the generating step includes generating a message with an indication of a fault when the current percentage is greater than the upper second threshold percentage and the lower second threshold percentage.