Adaptive quiesce for efficient cross-host consistent CDP checkpoints
A disaster recovery system, including a target datastore for replicating data written to source datastores, and a checkpoint engine (i) for transmitting, at multiple times, quiesce commands to a plurality of host computers, each quiesce command including a timeout period that is adjusted at each of the multiple times, (ii) for determining, at each of the multiple times, whether acknowledgements indicating that a host has successfully stopped writing enterprise data to the source datastores, have been received from each of the host computers within the timeout period, (iii) for marking, at each of the multiple times, a cross-host checkpoint in the target datastore and reducing the timeout period for the quiesce commands at the next time, if the determining is affirmative, and (iv) for increasing, at each of the multiple times, the timeout period for the quiesce commands transmitted at the next time, if the determining is not affirmative.
1. A method for generating cross-host consistent checkpoints for disaster recovery systems, comprising:
establishing a first time out period;
transmitting write quiesce commands to each of a plurality of enterprise host computers, the write quiesce commands including the first timeout period, wherein each of the plurality of enterprise host computers are operative to write enterprise data to source datastores;
receiving responses from each of the plurality of enterprise host computers, wherein each response indicates if a respective enterprise host computer has successfully stopped writing the enterprise data to the source datastores within the first time out period; and
comparing a response time from each of the plurality of enterprise host computers to the first time out period; and
determining a second time out period based on the responses from at least one of the plurality of enterprise host computers.
2. The method of claim 1 , comprising:
determining the value of the first time out period based on a production constraint.
3. The method of claim 2 wherein the production constraint limits a percentage of total time allocated for at least one quiesce timeout.
4. The method of claim 2 wherein the production constraint limits a number of quiesce timeouts that occur in a designated time period.
5. The method of claim 1 , comprising:
logging the write quiesce commands to write the enterprise data to the source datastores, issued by one or more of the plurality of enterprise host computers, in at least one of a plurality of write journals of a target datastore, and promoting the logged write quiesce commands to replication disks of the target datastore at a plurality of times, and marking checkpoints in each of the plurality of write journals of the target datastore.
6. The method of claim 5 , comprising:
recovering the source datastores based on images of the replication disks and on the plurality of write journals and the marked checkpoints.
7. The method of claim 1 , comprising:
determining the responses have not been received from each of the plurality of enterprise host computers within the first timeout period.
8. The method of claim 7 wherein the second time out period is greater than the first time out period.
9. The method of claim 1 , comprising:
determining the responses have been received from each of the plurality of enterprise host computers within the first timeout period.
10. The method of claim 9 , comprising:
marking a cross-host checkpoint in a target datastore for the enterprise host computers, wherein the second time period is less than the first time out period.
11. A disaster recovery system for providing continuous data protection to an enterprise, the enterprise comprising a plurality of host computers that are operative to read enterprise data from source datastores and write the enterprise data to the source datastores, comprising:
a target datastore that replicates the enterprise data written to the source datastores; and
a checkpoint engine that (i) establishes a first time out period; (ii) transmits write quiesce commands to each of a plurality of enterprise host computers, the write quiesce commands including the first time out period; (iii) receives responses from each of the plurality of enterprise host computers, wherein each response indicates whether or not a respective enterprise host computer successfully stopped writing the enterprise data to the source datastores within the first time out period; (iv) compares a response time from each of the plurality of enterprise host computers to the first time out period; and (v) determines a second time out period based on the responses from each of the plurality of enterprise host computers.
12. The system of claim 11 wherein said checkpoint engine determines the first time out period based on a production constraint.
13. The system of claim 11 wherein the production constraint limits a percentage of total time allocated for quiesce timeouts.
14. The system of claim 11 wherein the production constraint limits a number of quiesce timeouts that occur in a designated time period.
15. The system of claim 11 wherein the target datastore comprises:
a plurality of data disks for replication of the enterprise data in the source datastores; and
a plurality of write journals for logging the write quiesce commands issued by one or more of the plurality of enterprise host computers for writing the enterprise data into the source datastores, and wherein the checkpoint engine marks a cross-host checkpoint by marking checkpoints in each of the plurality of write journals of the target datastore.
16. The system of claim 15 further comprising a data recovery engine that recovers the source datastores based on images of the replication disks and based on the plurality of write journals and the checkpoints marked by the checkpoint engine.
17. The system of claim 11 , wherein the checkpoint engine determines that acknowledgements indicating that at least one of the plurality of enterprise host computers successfully stopped writing enterprise data to the source datastores have not been received from each of the plurality of enterprise host computers within the first timeout period.
18. The system of claim 17 , wherein the second time out period is greater than the first time out period.
19. The system of claim 11 wherein the checkpoint engine determines that acknowledgements indicating that at least one of the plurality of enterprise host computers successfully stopped writing enterprise data to the source datastores have been received from each of the plurality of enterprise host computers within the first timeout period.
20. The system of claim 19 , wherein the checkpoint engine marks marking a cross-host checkpoint in a target datastore for the enterprise host computer computers, and wherein the second time out period is less than the first time out period.