Method and apparatus for performing transparent in-memory checkpointing
View Patent ↗A method and apparatus for performing transparent checkpoint data backup by identifying portions of memory that store the checkpoint data that have been modified since a prior in-memory checkpoint process and backing up only the identified portions of memory containing the modified checkpoint data.
1. A method comprising:
accessing checkpoint data;
mapping the checkpoint data to a memory region;
initiating a backup of the checkpoint data, wherein
the initiating is performed after an initial backup of the checkpoint data has been performed;
in response to the initiating, identifying data stored in at least one portion of the memory region, wherein
the data stored in the at least one portion of the memory region has been modified since the initial backup of the checkpoint data;
copying the data stored in the at least one portion of the memory region; and
replicating the data stored in the at least one portion of the memory region to at least one backup server.
2. The method of claim 1 , wherein the at least one portion of the memory region comprises a memory page.
3. The method of claim 1 , wherein the at least one portion of the memory region comprises a memory block.
4. The method of claim 1 , further comprising:
monitoring one or more parameters associated with the memory region, wherein
the one or more parameters indicate whether a portion of the memory region has been modified.
5. The method of claim 1 , further comprising:
monitoring page-dirty bits to identify modified portions of the memory region.
6. The method of claim 1 , further comprising:
setting all portions of the memory region to read only;
tracking faults that occur each time a write request is executed for a portion of the memory region; and
identifying portions having faults as modified portions.
7. The method of claim 1 , further comprising:
tracking portions of memory that are being flushed by an operating system; and
identifying the portions as modified portions.
8. The method of claim 1 , wherein the memory region is at least one of a file or an operating system device.
9. A system comprising:
one or more processors;
memory coupled to the one or more processors, wherein the memory stores program instructions executable by the one or more processors to:
access checkpoint data,
map the checkpoint data to a memory region,
initiate a backup of the checkpoint data, wherein
the backup of the checkpoint data is initiated after an initial backup of the checkpoint data has been performed,
identify data stored in at least one portion of the memory region, in response to initiating the backup of the checkpoint data, wherein
the data stored in the at least one portion of the memory region has been modified since the initial backup of the checkpoint data,
copy the data stored in the at least one portion of the memory region, and
replicate the data stored in the at least one portion of the memory region to at least one backup server.
10. The system of claim 9 , wherein the at least one portion of the memory region comprises at least one of a memory page and a memory block.
11. The system of claim 9 , wherein the program instructions are further executable to:
monitor one or more parameters associated with the memory region, wherein
the one or more parameters indicate whether a portion of the memory region has been modified.
12. The system of claim 9 , wherein the program instructions are further executable to:
monitor page-dirty bits to identify modified portions of the memory region.
13. The system of claim 9 , wherein the program instructions are further executable to:
set all portions of the memory region to read only;
track faults that occur each time a write request is executed for a portion of the memory region; and
identify portions having faults as modified portions.
14. The system of claim 9 , wherein the program instructions are further executable to:
track portions of memory that are being flushed by an operating system; and
identify the portions as modified portions.
15. A computer readable storage medium comprising program instructions executable to:
access checkpoint data;
map the checkpoint data to a memory region;
initiate a backup of the checkpoint data, wherein
the backup of the checkpoint data is initiated after an initial backup of the checkpoint data has been performed;
identify data stored in at least one portion of the memory region, in response to
initiating the backup of the checkpoint data, wherein
the data stored in the at least one portion of the memory region has been modified since the initial backup of the checkpoint data;
copy the data stored in the at least one portion of the memory region; and
replicate the data stored in the at least one portion of the memory region to at least one backup server.
16. The computer readable storage medium of claim 15 , wherein the at least one portion of the memory region comprises at least one of a memory page and a memory block.
17. The computer readable storage medium of claim 15 , wherein the program instructions are further executable to:
monitor one or more parameters associated with the memory region, wherein
the one or more parameters indicate whether a portion of the memory region has been modified.
18. The computer readable storage medium of claim 15 , wherein the program instructions are further executable to:
monitor page-dirty bits to identify modified portions of the memory region.
19. The computer readable storage medium of claim 15 , wherein the program instructions are further executable to:
set all portions of the memory region to read only;
track faults that occur each time a write request is executed for a portion of the memory region; and
identify portions having faults as modified portions.
20. The computer readable storage medium of claim 15 , wherein the program instructions are further executable to:
track portions of memory that are being flushed by an operating system; and
identify the portions as modified portions.
21. A method comprising:
accessing checkpoint data;
mapping the checkpoint data to a memory region;
determining at least one portion of the memory region that has been modified;
setting all portions of the memory region to read only;
tracking faults that occur each time a write request is executed for a portion of the memory region;
identifying the portions having faults as portions that are modified portions;
copying the modified portions; and
replicating the copied modified portions of the memory region comprising modified checkpoint data to at least one backup server.