Enhanced operating system integrity using non-volatile system memory

• US 9,767,015 B1
• Filed: 11/01/2013
• Issued: 09/19/2017
• Est. Priority Date: 11/01/2013
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: McKelvie, Samuel James, Gupta, Anurag Windlass
• Primary Examiner(s): Bertram, Ryan
• Assistant Examiner(s): BLUST, JASON W
• Application Number:
  US14/070,350
• Time in Patent Office: 1,418 Days
• Field of Search: None
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 12/0238 Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory
• CPC Class Codes: G06F 12/08 in hierarchically structured memory systems, e.g. virtual memory systems
• 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 12/0897 with two or more cache hierarchy levels with multilevel cache hierarchies G06F12/0811
• CPC Class Codes: G06F 2212/1032 Reliability improvement, data loss prevention, degraded operation etc
• CPC Class Codes: G06F 2212/205 Hybrid memory, e.g. using both volatile and non-volatile memory
• CPC Class Codes: G06F 2212/463 File
• CPC Class Codes: G06F 2212/502 using adaptive policy
• CPC Class Codes: G06F 3/0607 by facilitating the process of upgrading existing storage systems, e.g. for improving compatibility between host and storage device
• CPC Class Codes: G06F 3/0655 Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices

Abstract

Methods and apparatus for enhancing operating system integrity using non-volatile system memory are described. A computer system includes a system memory coupled to one or more processors. The system memory comprises at least a non-volatile portion. Memory is allocated from the non-volatile portion to store selected metadata associated with an operating system component that supports input-output (I/O) operations. In response to an operation that results in a metadata change at the component, a metadata entry is stored in the non-volatile portion. Subsequent to a failure event, contents of the metadata entry are read from the non-volatile portion to restore a state of the component.

Claims

1. A system, comprising:

one or more processors, and

a system memory linked to the one or more processors, wherein the system memory comprises one or more non-volatile memory modules and one or more volatile memory modules;

a persistent storage device having a higher access latency than the system memory;

wherein the system memory comprises program instructions that when executed on the one or more processors;

identify, by an operating system, at least a selected subset of metadata associated with a state of a particular component of a plurality of components of the operating system, wherein the particular component of the operating system is configured to support input-output (I/O) operations;

designate, by the operating system, a portion of memory of a non-volatile memory module of the one or more non-volatile memory modules of the system memory to store at least the selected subset of metadata associated with the state of the particular component of the operating system;

designate, by the operating system, a portion of the memory of a volatile memory module of the one or more volatile memory modules of the system memory to store a non-selected subset of the metadata associated with the state of the particular component of the operating system;

in response to an operation that results in a change of the metadata associated with the state of the particular component of the operating system to a new state, store corresponding metadata entries for the new state of the particular component of the operating system only in the designated portions of the system memory;

retrieve, subsequent to a failure event that results in a loss of the non-selected subset of the metadata associated with a new state of the particular component of the operating system, the selected subset of the metadata associated with the new state of the particular component of the operating system stored in the designated portion of the non-volatile module of the system memory; and

restore, subsequent to the failure event, the state of the particular component of the operating system to the new state using the retrieved selected subset of the metadata.

2. The system as recited in claim 1, wherein the one or more non-volatile modules of the system memory comprise a non-volatile dual inline memory module (NV-DIMM).

3. The system as recited in claim 1, wherein the particular component comprises a file system management component, and wherein the selected subset of metadata associated with the new state comprises a portion of a file system metadata journal.

4. The system as recited in claim 1, wherein the particular component comprises a file management component configured to implement a write cache, wherein the program instructions when executed on the one or more processors:

store, within the portion of memory designated from the non-volatile memory module, data modified by a write operation directed at a particular file in use at the operating system.

5. The system as recited in claim 1, wherein the particular component comprises a driver for a block-level storage volume configured to implement a write cache, wherein the program instructions when executed on the one or more processors:

store, within the portion of memory designated from the non-volatile memory module, data modified by a write operation directed at a particular block-level storage volume in use at the operating system.

6. A method, comprising:

performing, by one or more computing devices comprising a system memory and a persistent storage device, wherein the system memory comprises one or more non-volatile memory modules and one or more volatile memory modules, and the persistent storage device comprises non-volatile memory having a higher access latency than the system memory;

identifying, by an operating system, metadata associated with a state of a particular component of a plurality of components of the operating system, wherein the particular component of the operating system is configured to support input-output (I/O) operations;

identifying, by the operating system, at least a selected subset of the metadata associated with a state of the particular component of the operating system;

allocating, by the operating system, a portion of memory of a non-volatile memory module of the one or more non-volatile memory modules of the system memory to store at least the selected subset of the metadata associated with the state of the particular component of the operating system;

allocating, by the operating system, a portion of the memory of a volatile memory module of the one or more volatile memory modules of the system memory to store the non-selected subset of the metadata associated with the state of the particular component of the operating system;

in response to an operation that results in a change of the metadata associated with the state of the particular component of the operating system to a new state,

storing, by the operating system, corresponding metadata entries for the new state of the particular component of the operating system only in the designated portion of the system memory;

retrieving, subsequent to a failure event that results in a loss of the non-selected subset of the metadata associated with the new state of the particular component of the operating system, the selected subset of the metadata associated with the new state of the particular component of the operating system stored in the designated portion of the non-volatile memory module of the system memory; and

restoring, subsequent to the failure event, the state of the particular component of the operating system to the new state using the retrieved selected subset of the metadata.

7. The method as recited in claim 6, wherein the non-volatile portion of the system memory comprises resistive memory.

8. The method as recited in claim 6, wherein the non-volatile portion of the system memory comprises phase-change memory.

9. The method as recited in claim 6, wherein the particular component comprises a file system management component, and wherein the selected subset of metadata comprises a portion of a file system metadata journal.

10. The method as recited in claim 6, wherein the particular component comprises a file management component configured to implement a write cache, further comprising performing, by the one or more computing devices:

storing, within the memory allocated from the non-volatile portion, data modified by a write operation directed at a particular file in use at the operating system.

11. The method as recited in claim 6, wherein the particular component comprises a driver for a block-level storage volume configured to implement a write cache, further comprising performing, by the one or more computing devices:

storing, within the memory allocated from the non-volatile portion, data modified by a write operation directed at a particular block-level storage volume in use at the operating system.

12. The method as recited in claim 6, further comprising performing, by the one or more computing devices:

configuring a portion of a storage device as a backing store for the selected set of the metadata.

13. The method as recited in claim 12, further comprising performing, by the one or more computing devices:

scheduling a consolidated write operation from the non-volatile portion of the system memory to the backing store, the consolidated write operation comprising the corresponding metadata entry and at least one other metadata entry.

14. The method as recited in claim 12, further comprising performing, by the one or more computing devices:

determining, based on one or more criteria, that a write of the metadata entry of the selected subset of the metadata associated with the state of the particular component of the operating system from the non-volatile portion of the system memory to the backing store is unnecessary; and

deleting the metadata entry from the non-volatile portion of the system memory without writing the metadata entry to the backing store.

15. The method as recited in claim 6, further comprising performing, by the one or more computing devices:

disabling, subsequent to the failure event, write access to the non-volatile portion of system memory until the state of the particular component of the operating system has been restored.

16. A non-transitory computer-accessible storage medium storing program instructions to execute on one or more computing devices comprising one or more processors that are linked to system memory, the system memory comprising one or more non-volatile memory modules and one or more volatile memory modules, the system memory having a lower access latency than a persistent storage device of the system, the program instructions executable on the one or more computing devices to:

identify, by an operating system, metadata associated with a state of a particular component of a plurality of components of the operating system, wherein the particular component of the operating system is configured to support input-output (I/O) operations;

identify, by the operating system, at least a selected subset of the metadata associated with a state of the particular component of the operating system;

designate, by the operating system, a portion of memory of a non-volatile memory module of the one or more non-volatile memory modules of the system memory to store at least the selected set of the metadata associated with a state of the particular component of the operating system;

designate, by the operating system, a portion of the memory of a volatile memory module of the one or more volatile memory modules of the system memory to store the non-selected subset of the metadata associated with the state of the particular component of the operating system;

in response to an operation that results in a change of the metadata associated with the state of the particular component of the operating system to a new state, store, by the operating system, corresponding metadata entries for the new state of the particular component of the operating system only in the designated portions of the system memory;

retrieve, subsequent to a failure event that results in a loss of the non-selected subset of the metadata associated with the new state of the particular component of the operating system, the selected subset of the metadata associated with the new state of the particular component of the operating system stored in the designated portion of the non-volatile module of the system memory; and

restore, subsequent to the failure event, the state of the particular component of the operating system to the new state using the retrieved selected subset of the metadata.

17. The non-transitory computer-accessible storage medium as recited in claim 16, wherein the non-volatile portion of the system memory comprises racetrack memory.

18. The non-transitory computer-accessible storage medium as recited in claim 16, wherein the particular component comprises a file system management component, and wherein the selected subset of metadata comprises a portion of a file system metadata journal.

19. The non-transitory computer-accessible storage medium as recited in claim 16, wherein the particular component comprises a file management component configured to implement a write cache, wherein the instructions when executed on the one or more processors:

store, within the memory designated from the non-volatile portion, data modified by a write operation directed at a particular file in use at the operating system.

20. The non-transitory computer-accessible storage medium as recited in claim 16, wherein the particular component comprises a driver for a block-level storage volume configured to implement a write cache, wherein the instructions when executed on the one or more processors:

store, within the memory designated from the non-volatile portion, data modified by a write operation directed at a particular block-level storage volume in use at the operating system.