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 use in disaster recovery systems, comprising:
transmitting, at multiple times, write quiesce commands to each of a plurality of enterprise host computers, each quiesce command including a timeout period that is adjusted at each of the multiple times, wherein the host computers are operative to write enterprise data to the source datastores;
at each of the multiple times:
determining whether acknowledgements indicating that a host has successfully stopped writing enterprise data to the source datastores, have been received from each of the plurality of host computers within the timeout period;
if said determining is affirmative, then marking a cross-host checkpoint in a target datastore for the host computers, and reducing the timeout period for the quiesce commands transmitted by said transmitting at the next time, wherein the target datastore is generated by a continuous data protection disaster recovery system to replicate the data in the source datastores; and
otherwise, increasing the timeout period for the quiesce commands transmitted by said transmitting at the next time.
2. The method of claim 1 further comprising, at each of the multiple times, dynamically determining the value of the next time, based on the current timeout period and based on a production constraint.
3. The method of claim 2 wherein the production constraint limits the percentage of total time allocated for quiesce timeouts.
4. The method of claim 2 wherein the production constraint limits the number of quiesce timeouts that occur in a designated time period.
5. The method of claim 1 further comprising logging each command to write enterprise data to a source datastore, issued by a host computer, in a write journal of the target datastore, and promoting the logged write commands to replication disks of the target datastore at a plurality of times, and wherein said marking a cross-host checkpoint comprises marking checkpoints in each of the write journals of the target datastore.
6. The method of claim 5 further comprising recovering the source datastores based on images of the replication disks and on the write journals and the checkpoints marked therein.
7. 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 enterprise data to source datastores, comprising:
a target datastore for replicating data written to the source datastores; and
a checkpoint engine (i) for transmitting, at multiple times, quiesce commands to each of the 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 plurality of host computers within the timeout period, (iii) for marking, at each of the multiple times, a cross-host checkpoint in said 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.
8. The system of claim 7 wherein said checkpoint engine dynamically determines, at each of the multiple times, the value of the next time, based on the current timeout period and based on a production constraint.
9. The system of claim 7 wherein the production constraint limits the percentage of total time allocated for quiesce timeouts.
10. The system of claim 7 wherein the production constraint limits the number of quiesce timeouts that occur in a designated time period.
11. The system of claim 7 wherein said target datastore comprises:
a plurality of data disks for replication of the data in the source datastores; and
a plurality of write journals for logging write commands issued by the host computers for writing data into the source datastores, and wherein said checkpoint engine marks a cross-host checkpoint by marking checkpoints in each of said write journals of said target datastore.
12. The system of claim 11 further comprising a data recovery engine for recovering the source datastores based on images of said replication disks and based on said write journals and the checkpoints marked therein by said checkpoint engine.