Balancing response times for synchronous I/O requests having different priorities
View Patent ↗A computing environment, such as an data mirroring or replication storage system, may need to process synchronous I/O requests having different priorities in addition to handling I/O requests on the basis of synchronous or asynchronous groupings. The system described herein provides a data storage system that addresses issues involving efficient balancing of response times for servicing synchronous I/O requests having different priorities. Accordingly, the system described herein provides for maintaining an optimal response time for the host-synchronous I/O requests and the optimal throughput of non-host-synchronous I/O requests using a host-synchronous request time window within which processing of non-host-synchronous I/O requests is throttled. The host-synchronous request time window may be selected to enable the optimal response time for the host-synchronous I/O and also to minimize the impact on the overall throughput of the I/O processor of the storage device.
1. A method of balancing response times of a system, comprising:
receiving a first-type synchronous I/O request from a host;
after receiving the first-type synchronous I/O request, initiating a time window having a length of time based on an assessment of an expected likelihood of receiving another first-type synchronous I/O request from the host;
in response to receiving at least one other first-type synchronous I/O request from the host during the time window, processing the at least one other first-type synchronous I/O request; and
in response to determining that at least one subsequent I/O request is a second-type request that is different from the at least one other first-type synchronous I/O request, throttling processing of the second-type request that is different from the at least one other first-type synchronous I/O request received during the time window until after the time window has expired.
2. The method according to claim 1 , wherein throttling the processing of the second-type request includes limiting processing resources available to process the second-type request.
3. The method according to claim 1 , wherein throttling the processing of the second-type request includes limiting generation of the second-type request.
4. The method according to claim 1 , wherein the second-type request is a synchronous I/O request having a different priority than the first-type synchronous I/O request.
5. The method according to claim 1 , wherein determining the length of the time window includes automatically determining the length based on at least one condition.
6. The method according to claim 5 , wherein the at least one condition includes at least one of: a network condition, a time of day, or an I/O request pattern.
7. The method according to claim 1 , wherein the length of the time window is fixed by a user.
8. A non-transitory computer-readable medium storing software for balancing response times of a system, the software comprising:
executable code that receives a first-type synchronous I/O request from a host;
executable code that, after receiving the first-type synchronous I/O request, initiates a time window having a length of time based on an assessment of an expected likelihood of receiving another first-type synchronous I/O request from the host;
executable code that, in response to receiving at least one other first-type synchronous I/O request from the host during the time window, processes the at least one other first-type synchronous I/O request; and
executable code that, in response to determining that at least one subsequent I/O request is a second-type request that is different from the at least one other first-type synchronous I/O request, throttles processing of the second-type request that is different from the at least one other first-type synchronous I/O request received during the time window until after the time window has expired.
9. The non-transitory computer-readable medium according to claim 8 , wherein the executable code that throttles the processing of the second-type request includes executable code that limits processing resources available to process the second-type request.
10. The non-transitory computer-readable medium according to claim 8 , wherein the executable code that throttles the processing of the second-type request includes executable code that limits generation of the second-type request.
11. The non-transitory computer-readable medium according to claim 8 , wherein the second-type request is a synchronous I/O request having a different priority than the host-synchronous I/O request.
12. The non-transitory computer-readable medium according to claim 8 , wherein the executable code that determines the length of the time window includes executable code that automatically determines the length based on at least one condition.
13. The non-transitory computer-readable medium according to claim 12 , wherein the at least one condition includes at least one of: a network condition, a time of day, or an I/O request pattern.
14. The non-transitory computer-readable medium according to claim 8 , wherein the executable code that determines the length of the time window includes executable code that provides an interface that receives input by a user for the length of the time window.
15. A system for balancing response times of a storage system, comprising:
at least one storage device that receives a first-type synchronous I/O request from a host; and
a throttle device that, after receiving the first-type synchronous I/O request, initiates a time window having a length of time based on an assessment of an expected likelihood of receiving another first-type synchronous I/O request from the host, wherein, in response to receiving at least one other first-type synchronous I/O request from the host during the time window, the throttle device enables processing of the at least one other first-type synchronous I/O request by the at least one storage device, and wherein in response to determining that at least one subsequent I/O request is a second-type request that is different from the at least one other first-type synchronous I/O request, the throttle device throttles processing of the second-type request that is different from the at least one other first-type synchronous I/O request received during the time window until after the time window has expired.
16. The system according to claim 1 , further comprising:
at least one remote storage device that is coupled to the at least one storage device and that provides data replication capability.
17. The system according to claim 15 , wherein the second-type request is a synchronous I/O request having a different priority than the first-type synchronous I/O request, and wherein throttling the processing of the second-type request includes at least one of: (i) limiting processing resources available to process the second-type request, or (ii) limiting generation of the second-type request.