Intelligent and automatic load balancing of workloads on replication appliances based on appliance load scores

1. A computer-implemented method comprising:

during a replication process between a primary site and a recovery site,

for each appliance of a plurality of appliances,

evaluating a replication load of the each appliance, wherein

the recovery site comprises the plurality of appliances,

the evaluating comprises

 analyzing at least a first evaluation factor for the each appliance and a second evaluation factor for the each appliance,

the first evaluation factor for the each appliance and the second evaluation factor for the each appliance are related to the replication load of the replication process on the each appliance,

the first evaluation factor is an evaluation factor for the each appliance,

the second evaluation factor is an evaluation factor for the replication load of the each appliance,

the each appliance of the plurality of appliances is a gateway appliance of a plurality of gateway appliances, and

the first evaluation factor and the second evaluation factor are ones of a plurality of evaluation factors,

determining a maximum load for the each appliance, and

determining an effective replication load score of a plurality of effective replication load scores, wherein

the effective replication load score is for the each appliance;

determining whether at least one appliance of the plurality of appliances is not overloaded based, at least in part, on the plurality of effective replication load scores; and

in response to a determination that the at least one appliance of the plurality of appliances is not overloaded,

sorting the plurality of effective replication load scores,

determining whether a relative difference exceeds a difference threshold, wherein

the relative difference is between a highest effective replication load score of the plurality of effective replication load scores and a lowest effective replication load score of the plurality of effective replication load scores, and

in response to a determination that the relative difference exceeds the difference threshold,

rebalancing one or more replication workloads between a plurality of the plurality of appliances.

2. The computer-implemented method of claim 1 , wherein

the method further comprises

determining a percentage of threshold value of a plurality of percentage of threshold values for the each of the plurality of appliances.

3. The computer-implemented method of claim 2 , further comprising:

applying a first weight to the first evaluation factor of the gateway appliance, and

applying a second weight to the second evaluation factor of the gateway appliance, wherein the first weight of the first evaluation factor of the gateway appliance is greater than the second weight of the second evaluation factor of the gateway appliance.

4. The computer-implemented method of claim 2 , wherein

at least one of the first evaluation factor and the second evaluation factor pertains to a current replication lag time associated with the gateway appliance.

5. The computer-implemented method of claim 2 , wherein

at least one of the first evaluation factor and the second evaluation factor is evaluated based on historic data pertaining to memory utilization of the gateway appliance.

6. The computer-implemented method of claim 2 , wherein

at least one of the first evaluation factor and the second evaluation factor pertains to a number of average unique blocks of data directed to the gateway appliance.

7. The computer-implemented method of claim 2 , wherein

determining whether at least one gateway appliance of the plurality of gateway appliances is overloaded; and

in response to a determination that at least one gateway appliance of the plurality of gateway appliances is overloaded,

performing another rebalancing operation on one or more of the plurality of gateway appliances based on one or more replication loads on one or more of the plurality of gateway appliances.

8. A system comprising:

one or more processing units, each including one or more processors and memory coupled to the one or more processors, wherein the memory stores program instructions executable by the one or more processors and the one or more processing units, alone or in combination with one another, execute the program instructions to perform a method comprising

during a replication process between a primary site and a recovery site,

for each appliance of a plurality of appliances,

evaluating a replication load of the each appliance, wherein

the recovery site comprises the plurality of appliances,

the evaluating comprises

 analyzing at least a first evaluation factor for the each appliance and a second evaluation factor for the each appliance, and

the first evaluation factor for the each appliance and the second evaluation factor for the each appliance are related to the replication load of the replication process on the each appliance,

the first evaluation factor is an evaluation factor for the each appliance,

the second evaluation factor is an evaluation factor for the replication load of the each appliance,

the each appliance of the plurality of appliances is a gateway appliance of a plurality of gateway appliances, and

the first evaluation factor and the second evaluation factor are ones of a plurality of evaluation factors,

determining a maximum load for the each appliance, and

determining an effective replication load score of a plurality of effective replication load scores, wherein

the effective replication load score is for the each appliance,

determining whether at least one appliance of the plurality of appliances is not overloaded based, at least in part, on the plurality of effective replication load scores, and

in response to a determination that the at least one appliance of the plurality of appliances is not overloaded,

sorting the plurality of effective replication load scores,

determining whether a relative difference exceeds a difference threshold, wherein

the relative difference is between a highest effective replication load score of the plurality of effective replication load scores and a lowest effective replication load score of the plurality of effective replication load scores, and

in response to a determination that the relative difference exceeds the difference threshold,

rebalancing one or more replication workloads between a plurality of the plurality of appliances.

9. The system of claim 8 , wherein

the method further comprises

determining a percentage of threshold value of a plurality of percentage of threshold values for the each of the plurality of appliances, and

in response to a determination that the relative difference in replication load does not exceed the difference threshold,

determining whether any gateway appliances of the plurality of gateway appliances has faulted, and

in response to a determination that a gateway appliance has faulted,

performing another rebalancing operation on one or more replication workloads between two or more gateway appliances of the plurality of gateway appliances that are online, and/or bringing one or more additional gateway appliances on line as part of the plurality of gateway appliances.

10. The system of claim 8 , wherein the method further comprises:

applying a first weight to the first evaluation factor of the gateway appliance, and

applying a second weight to the second evaluation factor of the gateway appliance, wherein the first weight of the first evaluation factor of the gateway appliance is greater than the second weight of the second evaluation factor of the gateway appliance.

11. The system of claim 8 , wherein

at least one of the first evaluation factor and the second evaluation factor pertains to a current replication lag time associated with the gateway appliance.

12. The system of claim 8 , wherein

at least one of the first evaluation factor and the second evaluation factor is evaluated based on historic data pertaining to memory utilization of the gateway appliance.

13. The system of claim 8 , wherein

at least one of the first evaluation factor and the second evaluation factor pertains to a number of average unique blocks of data directed to the gateway appliance.

14. The system of claim 8 , wherein the method further comprises:

determining whether at least one gateway appliance of the plurality of gateway appliances is overloaded; and

in response to a determination that at least one gateway appliance of the plurality of gateway appliances is overloaded,

performing another rebalancing operation on one or more of the plurality of gateway appliances based on one or more replication loads on the one or more of the plurality of gateway appliances.

15. A non-transitory computer-readable storage medium storing instructions executable by a processor to perform a method comprising:

during a replication process between a primary site and a recovery site,

for each appliance of a plurality of appliances,

evaluating a replication load of the each appliance, wherein

the recovery site comprises the plurality of appliances,

the evaluating comprises

 analyzing at least a first evaluation factor for the each appliance and a second evaluation factor for the each appliance, and

the first evaluation factor for the each appliance and the second evaluation factor for the each appliance are related to the replication load of the replication process on the each appliance,

the first evaluation factor is an evaluation factor for the each appliance,

the second evaluation factor is an evaluation factor for the replication load of the each appliance,

the each appliance of the plurality of appliances is a gateway appliance of a plurality of gateway appliances, and

the first evaluation factor and the second evaluation factor are ones of a plurality of evaluation factors,

determining a maximum load for the each appliance, and

determining an effective replication load score of a plurality of effective replication load scores, wherein

the effective replication load score is for the each appliance;

determining whether at least one appliance of the plurality of appliances is not overloaded based, at least in part, on the plurality of effective replication load scores; and

in response to a determination that the at least one appliance of the plurality of appliances is not overloaded,

sorting the plurality of effective replication load scores,

determining whether a relative difference exceeds a difference threshold, wherein

the relative difference is between a highest effective replication load score of the plurality of effective replication load scores and a lowest effective replication load score of the plurality of effective replication load scores, and

in response to a determination that the relative difference exceeds the difference threshold,

rebalancing one or more replication workloads between a plurality of the plurality of appliances.

16. The non-transitory computer-readable storage medium of claim 15 , wherein

the method further comprises

determining a percentage of threshold value of a plurality of percentage of threshold values for the each of the plurality of appliances, and

in response to a determination that the relative difference in replication load does not exceed the difference threshold,

determining whether any gateway appliances of the plurality of gateway appliances has faulted, and

in response to a determination that a gateway appliance has faulted,

rebalancing one or more replication workloads between two or more gateway appliances of the plurality of gateway appliances that are online, and/or bringing one or more additional gateway appliances on line as part of the plurality of gateway appliances.

17. The non-transitory computer-readable storage medium of claim 15 , wherein the method further comprises:

applying a first weight to the first evaluation factor of the gateway appliance, and

applying a second weight to the second evaluation factor of the gateway appliance, wherein the first weight of the first evaluation factor of the gateway appliance is greater than the second weight of the second evaluation factor of the gateway appliance.

18. The non-transitory computer-readable storage medium of claim 15 , wherein

at least one of the first evaluation factor and the second evaluation factor pertains to a current replication lag time associated with the gateway appliance.

19. The non-transitory computer-readable storage medium of claim 15 , wherein

at least one of the first evaluation factor and the second evaluation factor is evaluated based on historic data pertaining to memory utilization of the gateway appliance, and

at least one of the first evaluation factor and the second evaluation factor pertains to a number of average unique blocks of data directed to the gateway appliance.

20. The computer-implemented method of claim 2 , further comprising:

in response to a determination that the relative difference in replication load does not exceed the difference threshold,

determining whether any gateway appliances of the plurality of gateway appliances has faulted, and

in response to a determination that a gateway appliance has faulted,

performing another rebalancing operation on one or more replication workloads between two or more gateway appliances of the plurality of gateway appliances that are online, and/or bringing one or more additional gateway appliances on line as part of the plurality of gateway appliances.