Adaptive replication modes in a storage system
In one aspect, adaptive replication modes in a storage system are provided. An aspect includes during an active replication session in which a first type of replication is performed at the storage system, monitoring write input/output (IO) operations, collecting data from the write IO operations, and determining, from the collected data, write IO latency. Upon determining that a threshold value has been met from the write IO latency, where the threshold value is defined for the first type of replication, an aspect includes automatically switching from the first type of replication to a second type of replication. The second type of replication is configured to compensate for operational deficiencies detected in response to the write IO latency.
1. A method for adaptive replication in a storage system, the method comprising:
during an active replication session in which a first type of replication is performed at the storage system:
monitoring write input/output (IO) operations;
collecting data from the write IO operations;
determining, from the collected data, write IO latency, wherein the write IO latency for the first type of replication includes a length of time taken to mirror the write IO operations from a source cluster to a target cluster; and
upon determining that a threshold value has been met from the write IO latency, wherein the threshold value is defined for the first type of replication, automatically switching from the first type of replication to a second type of replication, wherein the second type of replication is configured to compensate for operational deficiencies detected in response to the write IO latency,
wherein the first type of replication comprises synchronous replication and the second type of replication comprises asynchronous replication, and wherein the threshold value is met when an average value of the write IO latency is persisted for a pre-defined amount of time.
2. The method of claim 1 , wherein the threshold value is met when the write IO latency for the first type of replication causes a consistency group trip and recovery after a failed consistency group trip.
3. The method of claim 1 , wherein automatically switching to the second type of replication includes:
selecting a most recent recovery snap set pair as a base;
creating identical snap set pairs based on the recovery snap set pair; and
initiating an asynchronous replication cycle.
4. The method of claim 1 , wherein the first type of replication comprises asynchronous replication and the second type of replication comprises synchronous replication, wherein the threshold value defines an average and peak latency and the threshold value is met when the write IO latency average and peak IO latency is below the threshold value.
5. The method of claim 4 , wherein automatically switching between the first type of replication to the second type of replication includes:
selecting a most recent synchronized snap set pair as a base; and
beginning synchronous replication IO mirroring concurrently with an async-to-sync transition cycle, wherein full synchronization is reached at completion of the async-to-sync transition cycle.
6. The method of claim 1 , wherein the threshold value for the first type of replication is a user-tunable value.
7. A system for adaptive replication in a storage system, the system comprising:
a memory comprising computer-executable instructions; and
a processor executing the computer-executable instructions, the computer-executable instructions when executed by the processor cause the processor to perform operations comprising:
during an active replication session in which a first type of replication is performed at the storage system:
monitoring write input/output (IO) operations;
collecting data from the write IO operations;
determining, from the collected data, write IO latency, wherein the write IO latency for the first type of replication includes a length of time taken to mirror the write IO operations from a source cluster to a target cluster; and
upon determining that a threshold value has been met from the write IO latency, wherein the threshold value is defined for the first type of replication, automatically switching from the first type of replication to a second type of replication, wherein the second type of replication is configured to compensate for operational deficiencies detected in response to the write IO latency,
wherein the first type of replication comprises synchronous replication and the second type of replication comprises asynchronous replication, and wherein the threshold value is met when an average value of the write IO latency is persisted for a pre-defined amount of time.
8. The system of claim 7 , wherein the threshold value is met when the write IO latency for the first type of replication causes a consistency group trip and recovery after a failed consistency group trip.
9. The system of claim 7 , wherein automatically switching to the second type of replication includes:
selecting a most recent recovery snap set pair as a base;
creating identical snap set pairs based on the recovery snap set pair; and
initiating an asynchronous replication cycle.
10. The system of claim 7 , wherein the first type of replication comprises asynchronous replication and the second type of replication comprises synchronous replication, wherein the threshold value defines an average and peak latency and the threshold value is met when the write IO latency average and peak IO latency is below the threshold value.
11. The system of claim 10 , wherein automatically switching between the first type of replication to the second type of replication includes:
selecting a most recent synchronized snap set pair as a base; and
beginning synchronous replication IO mirroring concurrently with an async-to-sync transition cycle, wherein full synchronization is reached at completion of the async-to-sync transition cycle.
12. A computer program product for adaptive replication in a storage system, the computer program product embodied on a non-transitory computer readable medium and including instructions that, when executed by a computer causes the computer to perform operations comprising:
monitoring write input/output (IO) operations;
collecting data from the write IO operations;
determining, from the collected data, write IO latency; and
upon determining that a threshold value has been met from the write IO latency, wherein the threshold value is defined for the first type of replication, automatically switching from the first type of replication to a second type of replication, wherein the second type of replication is configured to compensate for operational deficiencies detected in response to the write IO latency,
wherein automatically switching to the second type of replication includes:
selecting a most recent recovery snap set pair as a base;
creating identical snap set pairs based on the recovery snap set pair; and
initiating an asynchronous replication cycle.
13. The computer program product of claim 12 , wherein the first type of replication comprises synchronous replication and the second type of replication comprises asynchronous replication;
wherein the write IO latency for the first type of replication includes a length of time taken to mirror the write IO operations from a source cluster to a target cluster, and wherein the threshold value is met when an average value of the write IO latency is persisted for a pre-defined amount of time.
14. The computer program product of claim 12 , wherein the first type of replication comprises synchronous replication and the second type of replication comprises asynchronous replication; and
wherein the threshold value is met when the write IO latency for the first type of replication causes a consistency group trip and recovery after a failed consistency group trip.
15. The computer program product of claim 12 , wherein the first type of replication comprises asynchronous replication and the second type of replication comprises synchronous replication, wherein the threshold value defines an average and peak latency and the threshold value is met when the write IO latency average and peak IO latency is below the threshold value; and
wherein automatically switching between the first type of replication to the second type of replication includes:
selecting a most recent synchronized snap set pair as a base; and
beginning synchronous replication IO mirroring concurrently with an async-to-sync transition cycle, wherein full synchronization is reached at completion of the async-to-sync transition cycle.