Use of timestamp logic in synchronous replication
View Patent ↗Systems and methods for maintaining data consistency across mirrored storage nodes. Data consistency may be maintained in networked storage environments using a timestamp of an I/O in the primary and secondary servers (also referred to as nodes) in synchronous replication to re-synchronize the data efficiently, without much performance drop and while reducing the memory footprint. According to an aspect presented herein, data consistency is maintained in a mirror node by timestamping the I/O for a volume's logical territory in both the primary and secondary nodes. The timestamp acts as a watermark. Thus, whenever a disaster or failure occurs, the watermark point determines the re-synchronization point.
1. A method of using timestamps in a synchronous replication between a first volume and a second volume, comprising:
maintaining a global time segment number that is incremented in time;
logging a current global time segment number for each write to a provision with a territory within a redundant array of inexpensive disk (RAID) volume, the logging of the current global time segment number for each write being performed if a territory time segment number is less than the current global time segment number for each write being performed at a territory granularity;
determining a restoration of replication between the first volume and the second volume after a failure has been detected;
determining a last global time segment number where replication was successful and declaring the last global time segment as a watermark; and
restoring I/Os to the second volume from the first volume from that occurred since the watermark.
2. The method of claim 1 , wherein the global time segment number has a fixed length.
3. The method of claim 1 , wherein the logging is performed by a distributed volume manager (DVM) associated with the territory.
4. The method of claim 1 , further comprising:
declaring the watermark at a time corresponding to the last global time segment minus 1; and
restoring I/Os to the second volume from the watermark.
5. The method of claim 1 , further comprising clearing the watermark after all I/Os since the watermark have been successfully restored.
6. The method of claim 1 , further comprising:
restoring a portion of the I/Os since the watermark by beginning a resynchronization with territories having a lowest time segment number;
updating the territories on the second volume;
incrementing a time segment number being restored; and
updating the watermark with the incremented time segment number after the second volume is updated.
7. The method of claim 1 , wherein each global time segment represents 20 minutes.
8. The method of claim 1 , logging the current global time segment number for each write further comprising writing the current global time segment to a 3-byte field in a DVM territory metadata.
9. A data storage system, comprising:
a storage server;
a physical storage device associated with storage server;
a processing unit associated with the storage server; and
one or more modules for execution on the processing unit, operable to maintain a global time segment number that is incremented in time, log a current global time segment number for each write to a provision with a territory within a redundant array of inexpensive disk (RAID) volume, determine a restoration of replication between a first volume and a second volume after a failure has been detected, determine a last global time segment number where replication was successful and declaring the last global time segment as a watermark, and restore I/Os to the second volume from the first volume from that occurred since the watermark,
wherein the current global time segment number for each write is logged if a territory time segment number is less than the current global time segment number for each write being performed at a territory granularity.
10. The data storage system of claim 9 , wherein the global time segment number has a fixed length.
11. The data storage system of claim 9 , further comprising a distributed volume manager (DVM) associated with the territory that performs the logging.
12. The data storage system of claim 9 , wherein the watermark is declared at a time corresponding to the last global time segment minus 1, and wherein I/Os are restored to the second volume from the watermark.
13. The data storage system of claim 9 , wherein the watermark is cleared after all I/Os since the watermark have been successfully restored.
14. The data storage system of claim 9 , wherein a portion of the I/Os are restored since the watermark by beginning a resynchronization with territories having a lowest time segment number, wherein the territories are updated on the second volume, wherein a time segment number being restored is incremented, an wherein the watermark is updated with the incremented time segment number after the second volume is updated.
15. A computer storage medium having computer-executable instructions stored thereon which, when executed by a computer system, cause the computer system to:
maintain a global time segment number that is incremented in time;
log a current global time segment number for each write to a provision with a territory within a redundant array of inexpensive disk (RAID) volume;
determine a restoration of replication between a first volume and a second volume after a failure has been detected;
determine a last global time segment number where replication was successful and declaring the last global time segment as a watermark; and
restore I/Os to the second volume from the first volume from that occurred since the watermark,
wherein the current global time segment number for each write is logged if a territory time segment number is less than the current global time segment number for each write being performed at a territory granularity.
16. The computer storage medium of claim 15 , wherein the global time segment number has a fixed length.
17. The computer storage medium of claim 15 , wherein the logging is performed by a distributed volume manager (DVM) associated with the territory.
18. The computer storage medium of claim 15 , further comprising instructions to:
declare the watermark at a time corresponding to the last global time segment minus 1; and
restore I/Os to the second volume from the watermark.
19. The computer storage medium of claim 15 , further comprising instructions to clear the watermark after all I/Os since the watermark have been successfully restored.
20. The computer storage medium of claim 15 , further comprising instructions to:
restore a portion of the I/Os since the watermark by beginning a resynchronization with territories having a lowest time segment number;
update the territories on the second volume;
increment a time segment number being restored; and
update the watermark with the incremented time segment number after the second volume is updated.