System and method for hybrid kernel- and user-space checkpointing

• US 8,745,442 B1
• Filed: 04/28/2011
• Issued: 06/03/2014
• Est. Priority Date: 04/28/2011
• Status: Active Grant

• Current Assignee: Philips North America LLC (Koninklijke Philips N.V.)
• Original Assignee: Open Invention Network LLC
• Inventors: Havemose, Allan
• Primary Examiner(s): Lottich, Joshua P
• Application Number:
  US13/096,461
• Time in Patent Office: 1,132 Days
• Field of Search: 714/13, 714/15
• US Class Current: 714/15
• CPC Class Codes: G06F 11/1407 Checkpointing the instruction stream
• CPC Class Codes: G06F 11/1438 Restarting or rejuvenating
• CPC Class Codes: G06F 11/1446 Point-in-time backing up or restoration of persistent data
• CPC Class Codes: G06F 11/1451 by selection of backup contents
• CPC Class Codes: G06F 11/1469 Backup restoration techniques
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 2201/82 Solving problems relating to consistency
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data

Abstract

A system, method, and computer readable medium for hybrid kernel-mode and user-mode checkpointing of multi-process applications. The computer readable medium includes computer-executable instructions for execution by a processing system. A multi-process application runs on primary hosts and is checkpointed by a checkpointer comprised of a kernel-mode checkpointer module and one or more user-space interceptors providing barrier synchronization, checkpointing thread, resource flushing, and an application virtualization space. Checkpoints may be written to storage and the application restored from said stored checkpoint at a later time. Checkpointing is transparent to the application and requires no modification to the application, operating system, networking stack or libraries. In an alternate embodiment the kernel-mode checkpointer is built into the kernel.

Claims

1. A system for providing hybrid kernel-space and user-space checkpointing, the system comprising:

computer system memory comprising one or more memory locations configured to store one or more multi-process applications;

one or more Central Processing Units (CPUs) operatively connected to said computer system memory and configured to execute said one or more multi-process applications on a host with a host operating system;

one or more additional computer system memory or storage locations configured to store checkpoints;

one or more networking stacks; and

a hybrid kernel and user space checkpointer comprised of a kernel-space checkpoint device driver, and

one or more user-space interceptors preloaded into an address space of each application process, wherein said checkpointer is configured to checkpoint said one or more multi-process applications;

wherein said user-space interceptors comprise a user-space barrier and an application virtualization space providing a private resource name space; and

wherein state in said networking stacks is flushed as part of checkpointing said one or more multi-process applications.

2. The system according to claim 1 wherein said host operating system is one of Windows, Linux, Solaris or UNIX.

3. The system according to claim 1 wherein said kernel-space checkpoint device driver is implemented as one of a kernel module, a loadable kernel module, or kernel loadable module.

4. The system according to claim 1 wherein said kernel-space checkpoint device driver is compiled into a kernel.

5. The system according to claim 1 wherein said kernel-space checkpointer device driver is implemented as one of a character device or block device.

6. The system according to claim 5 wherein a checkpoint is taken by a call of a read function of said character device or block device, and the return value of said read function is the number of bytes checkpointed.

7. The system according to claim 6, wherein said read function calls are aligned to page boundaries of the host operating system.

8. The system according to claim 6, wherein said read function call is optimized by one or more of skipping reserved low memory, skipping reserved high memory, or skipping memory reserved for the operating system.

9. The system according to claim 6, wherein said read function calls returning zero are excluded from the checkpoint.

10. The system according to claim 5, wherein said character device or block device is accessed through a virtual file system (VFS).

11. The system according to claim 1, wherein data is transferred from kernel-space to user-space using copy_to_user( ) or memory mapped pages.

12. The system according to claim 11, wherein invalid pages are detected using access_ok( ) and skipped by said copy_to_user( ) function.

13. The system according to claim 9, further comprising flushing application resources while application execution is halted at said user-space barrier.

14. The system according to claim 1, wherein said kernel-space checkpointer device driver is called only with application execution halted at said user-space barrier.

15. The system according to claim 1, wherein said application virtualization space provides mappings between resources in an original application checkpoint and the corresponding application resources within an application restored from said original application checkpoint.

16. The system according to claim 15, wherein said mappings are provided for one or more of process IDs, thread IDs, and files.

17. The system according to claim 15, wherein said application virtualization space is implemented as one of a database, hashtable, or custom coding.

18. A non-transitory computer readable storage medium including a computer program having instructions for performing hybrid kernel-space and user-space checkpointing comprising:

computer system memory comprising one or more memory locations configured to store one or more multi-process applications;

one or more Central Processing Units (CPUs) operatively connected to said computer system memory and configured to execute said one or more multi-process applications on a host with a host operating system;

one or more additional computer system memory or storage locations configured to store checkpoints; and

a hybrid kernel and user space checkpointer comprised of a kernel-space checkpoint device driver, and

one or more user-space interceptors preloaded into an address space of each application process, wherein said checkpointer is configured to checkpoint said one or more multi-process applications;

wherein said user-space interceptors comprise a user-space barrier and an application virtualization space providing a private resource name space; and

wherein state in said networking stacks is flushed as part of checkpointing said one or more multi-process applications.

19. The non-transitory computer readable storage medium according to claim 18 wherein said kernel-space checkpoint device driver is implemented as one of a character device or block device.

20. The non-transitory computer readable storage medium according to claim 18 wherein said kernel-space checkpoint device driver is implemented as one of a kernel module, a loadable kernel module, kernel loadable module, or compiled into a kernel.