Multi-path layer configured to access storage-side performance metrics for load balancing policy control
An apparatus comprises at least one processing device. The at least one processing device is configured to obtain storage-side performance information maintained by a storage system in conjunction with processing of input-output operations directed to the storage system by a host device over a network, to dynamically select a particular one of a plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information, and to apply the selected load balancing policy in directing additional input-output operations from the host device to the storage system. At least one of the load balancing policies comprises a storage cache aware load balancing policy that causes different ones of the input-output operations to be directed to different cache entities of the storage system based at least in part on cache-related performance metrics of the obtained storage-side performance information.
1. An apparatus comprising:
at least one processing device comprising a processor coupled to a memory;
wherein the at least one processing device is configured:
to obtain storage-side performance information maintained by a storage system in conjunction with processing of input-output operations directed to the storage system by a host device over a network, the storage-side performance information comprising at least cache collision rates for respective cache entities of the storage system;
to dynamically select a particular one of a plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information in conjunction with host-side performance information maintained by the host device; and
to apply the selected load balancing policy in directing additional input-output operations from the host device to the storage system.
2. The apparatus of claim 1 wherein the at least one processing device comprises at least a portion of the host device.
3. The apparatus of claim 1 wherein the obtaining, dynamically selecting and applying are implemented in a multi-path layer comprising at least one multi-path input-output driver of the host device.
4. The apparatus of claim 1 wherein at least one of the load balancing policies comprises a storage cache aware load balancing policy that causes different ones of the input-output operations to be directed to different cache entities of the storage system based at least in part on cache-related performance metrics of the obtained storage-side performance information.
5. The apparatus of claim 4 wherein the cache entities of the storage system comprise respective components of the storage system that have respective local caches associated therewith.
6. The apparatus of claim 4 wherein the cache entities of the storage system comprise respective storage controllers of the storage system with each such storage controller having a different local cache associated therewith.
7. The apparatus of claim 1 wherein dynamically selecting a particular one of the plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information comprises automatically switching between at least a first load balancing policy that is storage cache aware and a second load balancing policy that is not storage cache aware.
8. The apparatus of claim 7 wherein the first load balancing policy utilizes cache-related performance metrics of the obtained storage-side performance information in load balancing decisions.
9. The apparatus of claim 7 wherein the second load balancing policy does not utilize cache-related performance metrics of the obtained storage-side performance information in load balancing decisions.
10. The apparatus of claim 7 wherein the first load balancing policy comprises:
accessing for each of one or more storage devices a stored mapping between ranges of logical block addresses of the storage device and respective ones of a plurality of cache entities of the storage system; and
for each of at least a subset of the additional input-output operations:
identifying a particular one of the cache entities based at least in part on a logical block address of the input-output operation and the stored mapping; and
selecting a particular one of a plurality of paths for delivery of the input-output operation to the storage system based at least in part on the identified cache entity.
11. The apparatus of claim 10 wherein the mapping is generated at least in part using information obtained from the storage system and characterizing the cache entities of the storage system.
12. The apparatus of claim 1 wherein the storage-side performance information is obtained from the storage system utilizing at least one of:
an in-band communication mechanism in which one or more commands in a designated storage protocol are sent from the host device to the storage system; and
an out-of-band communication mechanism in which host management software of the host device communicates with storage system management software of the storage system over a network connection.
13. The apparatus of claim 1 wherein dynamically selecting a particular one of the plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information comprises automatically selecting a first load balancing policy that is storage cache aware to replace a second load balancing policy that is not storage cache aware responsive to one or more of the cache collision rates for respective cache entities of the storage system exceeding a specified threshold.
14. A computer program product comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code, when executed by at least one processing device comprising a processor coupled to a memory, causes the processing device:
to obtain storage-side performance information maintained by a storage system in conjunction with processing of input-output operations directed to the storage system by a host device over a network, the storage-side performance information comprising at least cache collision rates for respective cache entities of the storage system;
to dynamically select a particular one of a plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information in conjunction with host-side performance information maintained by the host device; and
to apply the selected load balancing policy in directing additional input-output operations from the host device to the storage system.
15. The computer program product of claim 14 wherein at least one of the load balancing policies comprises a storage cache aware load balancing policy that causes different ones of the input-output operations to be directed to different cache entities of the storage system based at least in part on cache-related performance metrics of the obtained storage-side performance information.
16. The computer program product of claim 14 wherein dynamically selecting a particular one of the plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information comprises automatically switching between at least a first load balancing policy that is storage cache aware and a second load balancing policy that is not storage cache aware.
17. A method comprising:
obtaining storage-side performance information maintained by a storage system in conjunction with processing of input-output operations directed to the storage system by a host device over a network, the storage-side performance information comprising at least cache collision rates for respective cache entities of the storage system;
dynamically selecting a particular one of a plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information in conjunction with host-side performance information maintained by the host device; and
applying the selected load balancing policy in directing additional input-output operations from the host device to the storage system.
18. The method of claim 17 wherein at least one of the load balancing policies comprises a storage cache aware load balancing policy that causes different ones of the input-output operations to be directed to different cache entities of the storage system based at least in part on cache-related performance metrics of the obtained storage-side performance information.
19. The method of claim 17 wherein dynamically selecting a particular one of the plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information comprises automatically switching between at least a first load balancing policy that is storage cache aware and a second load balancing policy that is not storage cache aware.
20. The method of claim 17 wherein dynamically selecting a particular one of the plurality of distinct load balancing policies available in the host device based at least in part on the obtained storage-side performance information comprises automatically selecting a first load balancing policy that is storage cache aware to replace a second load balancing policy that is not storage cache aware responsive to one or more of the cache collision rates for respective cache entities of the storage system exceeding a specified threshold.