Back-up and restoration of data between volatile and flash memory

• US 9,632,715 B2
• Filed: 08/10/2015
• Issued: 04/25/2017
• Est. Priority Date: 08/10/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Palmer, Michael J., Wong, Kelvin
• Primary Examiner(s): Ruiz, Aracelis
• Application Number:
  US14/822,140
• Publication Number:
  US 20170046081A1
• Time in Patent Office: 624 Days
• CPC Class Codes: G01K 11/265 using surface acoustic wave [SAW]
• CPC Class Codes: G01K 2205/00 Application of thermometers in motors, e.g. of a vehicle
• CPC Class Codes: G01M 15/05 by combined monitoring of two or more different engine parameters
• CPC Class Codes: G01N 29/2481 Wireless probes, e.g. with transponders or radio links
• CPC Class Codes: G06F 12/0868 Data transfer between cache memory and other subsystems, e.g. storage devices or host systems
• CPC Class Codes: G06F 2212/1032 Reliability improvement, data loss prevention, degraded operation etc
• CPC Class Codes: G06F 2212/2022 Flash memory
• CPC Class Codes: G06F 2212/222 Non-volatile memory
• CPC Class Codes: G06F 2212/7209 Validity control, e.g. using flags, time stamps or sequence numbers
• 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 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0656 Data buffering arrangements
• 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
• CPC Class Codes: H04Q 9/00 Arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom

Abstract

A method and system are provided for back-up and restoration of data between volatile and flash memory. The method for controlling back-up of data to flash memory includes: organizing back-up data into stripes, wherein a stripe is a set of pages across all available flash memory devices, dies and planes which have the same block and page address; maintaining metadata indicating locations of known bad planes and grown bad planes; using the metadata when writing back-up data to determine which planes to send cache program commands to; and sending cache program commands to three or more stripes of data simultaneously including providing an indication in the stripe that the stripe is handling a cache program command. If a grown bad block is encountered while saving a stripe of data, the stripe of data is re-written to the next available page address avoiding the grown bad block.

Claims

1. A method for controlling back-up of data to flash memory, comprising:

organizing back-up data into stripes, wherein a stripe is a set of pages across all available flash memory devices, dies and planes which have the same block and page address;

maintaining metadata indicating locations of known bad planes and grown bad planes, wherein maintaining metadata includes maintaining a combined bad plane register of known bad planes and grown bad planes, wherein the combined bad plane register carries out an OR operation of the known bad planes and the grown bad planes;

using the metadata when writing back-up data to determine which planes to send cache program commands to; and

sending cache program commands to three or more stripes of data simultaneously including providing an indication in a stripe that the stripe is handling a cache program command;

wherein if a grown bad block is encountered whilst saving a stripe of data, the stripe of data is re-written to the next available page address avoiding the grown bad block.

2. The method as claimed in claim 1, wherein providing an indication in a stripe being written that the stripe is handling a cache program command includes setting a bit in metadata indicating that the following page in the plane will also be programmed.

3. The method as claimed in claim 1, including for each plane a cache register allowing a host to move data in or out of flash memory whilst the flash memory is busy carrying out an array operation.

4. The method as claimed in claim 3, wherein sending cache program commands to three or more stripes of data simultaneously includes allowing a host to transfer data for a second page program whilst a plane is busy completing the previous page program.

5. The method as claimed in claim 1, including for each stripe determining if any pages have failed;

if no pages have failed in a stripe, continue writing next stripe;

if one or more pages have failed in a stripe, update the grown bad plane metadata for the three or more stripes, end a cache program in next stripe, and re-write the previous three or more stripes.

6. The method as claimed in claim 1, wherein maintaining metadata indicating locations of known bad planes and grown bad planes includes a known bad plane register, a grown bad plane register, and a combined register of all bad planes.

7. The method as claimed in claim 1, including embedding bad block metadata in the backed-up data for use during restoration.

8. The method as claimed in claim 1, further comprising restoring backed-up data from flash memory including reading metadata from a stripe of backed-up data, and issuing cache read commands for planes for which the metadata is good.

9. The method as claimed in claim 8, wherein restoring backed-up data from flash memory includes updating the metadata in the form known bad planes and grown bad planes from metadata embedded in the backed-up data in the flash memory.

10. A system for controlling back-up of data to flash memory, wherein the system includes a back-up engine having a processor for backing-up data to flash memory formed of multiple flash devices, wherein the back-up engine provides commands to multiple flash device controllers having dedicated flash interfaces to the multiple flash devices, the back-up engine comprising:

a data structure component for organizing back-up data into stripes, wherein a stripe is a set of pages across all available flash memory devices, dies and planes which have the same block and page address;

a volatile storage for storing metadata regarding the back-up process including maintaining metadata indicating locations of known bad planes and grown bad planes, wherein the volatile storage has a temporary power supply, and wherein the volatile storage includes a combined bad plane register of known bad planes and grown bad planes, wherein the combined bad plane register carries out an OR operation of the known bad planes and the grown bad planes;

a back-up writing component using the metadata when writing back-up data to determine which planes to send cache program commands to;

a caching component for sending cache program commands to three or more stripes of data simultaneously including providing an indication in a stripe that the stripe is handling a cache program command; and

a re-writing component, wherein if a grown bad block is encountered whilst saving a stripe of data, the stripe of data is re-written to the next available page address avoiding the grown bad block.

11. The system as claimed in claim 10, wherein the caching component includes providing an indication in a stripe being written that the stripe is handling a cache program command and setting a bit in metadata indicating that the following page in the plane will also be programmed.

12. The system as claimed in claim 10, wherein the caching component includes, for each plane, a cache register allowing a host to move data in or out of flash memory whilst the flash memory is busy carrying out an array operation.

13. The system as claimed in claim 12, wherein the caching component configured to send cache program commands to three or more stripes of data simultaneously includes allowing a host to transfer data for a second page program whilst a plane is busy completing the previous page program.

14. The system as claimed in claim 10, the re-write component for each stripe determines if any pages have failed;

if no pages have failed in a stripe, the re-write component continues writing next stripe;

if one or more pages have failed in a stripe, the re-write component updates the grown bad plane metadata for three or more stripes, ends a cache program in next stripe, and re-writes the previous three or more stripes.

15. The system as claimed in claim 10, wherein the volatile storage for storing metadata indicating locations of known bad planes and grown bad planes includes a known bad plane register, a grown bad plane register, and a combined register of all bad planes.

16. The system as claimed in claim 10, including a metadata embedding component for embedding bad block metadata in the backed-up data for use during restoration.

17. The system as claimed in claim 10, further comprising a restoring component for restoring backed-up data from flash memory including reading metadata from a stripe of backed-up data, issuing cache read commands for planes for which the metadata is good.

18. The system as claimed in claim 17, wherein the restoring component includes updating the volatile storage to store the metadata read from the backed-up data including locations of known bad planes and grown bad planes.