Techniques for storage network bandwidth management
Techniques for storage network bandwidth management are disclosed. In one particular exemplary embodiment, the techniques may be realized as a method for storage network bandwidth management comprising sampling, using at least one computer processor, application Input/Output (I/O) requests associated with the unit of storage during a specified period of time, determining a maximum latency value based on the sampling of the application Input/Output (I/O) requests, comparing the maximum latency value with a current latency value, and throttling administrative I/O requests in the event that the current latency value exceeds the maximum latency value.
1. A method for storage network bandwidth management comprising:
sampling, using a storage network bandwidth management component, application Input/Output (I/O) requests associated with a first unit of storage during a specified period of time, wherein the sampling is independent of volume management administrative I/O requests;
determining a maximum latency value based on the sampling of the application Input/Output (I/O) requests;
comparing the maximum latency value with a current application I/O latency value;
throttling the volume management administrative I/O requests in the event that the current application I/O latency value exceeds the maximum latency value, wherein the volume management administrative I/O requests are independent of the application Input/Output (I/O) requests; and
increasing the volume management administrative Input/Output (I/O) requests in the event the current application I/O latency value is a specified level below the maximum latency value, wherein the volume management administrative I/O requests are independent of the application Input/Output (I/O) requests and wherein a level of the increase is based upon an amount the current application I/O latency level is below the maximum latency value.
2. The method of claim 1 , further comprising:
periodically sampling the application Input/Output (I/O) requests to determine a new maximum latency value.
3. The method of claim 2 , wherein the periodic sampling results in a plurality of stored maximum latency values used to identify bandwidth utilization trends.
4. The method of claim 1 , wherein the first unit of storage comprises at least one of a volume, a disk, and a disk array.
5. The method of claim 1 , wherein determining a maximum latency value based on the sampling of the application Input/Output (I/O) requests comprises specifying a maximum latency value based on a percentage of an average latency of the sampled application Input/Output (I/O) requests.
6. The method of claim 1 , wherein throttling volume management administrative I/O requests in the event that the current application I/O latency value exceeds the maximum latency value comprises decreasing the volume management administrative I/O requests at a rate proportional to a rate at which the current application I/O latency value exceeds the maximum latency value.
7. The method of claim 1 , further comprising:
comparing the current application I/O latency value with a maximum latency value associated with a second unit of storage, wherein throttling volume management administrative I/O requests comprises throttling the volume management administrative I/O requests in the event that the current application I/O latency value exceeds either the maximum latency value of the first unit of storage or the maximum latency value of the second unit of storage.
8. The method of claim 7 , wherein the comparison is based upon an administrative operation occurring which requires writing from the first unit of storage to the second unit of storage.
9. The method of claim 1 , wherein throttling volume management administrative I/O requests is implemented using a sliding window protocol.
10. The method of claim 1 , wherein I/O requests of an application are non-disruptively differentiated from the volume management administrative I/O requests based on a field in a packet header of volume management administrative I/O request packets.
11. At least one non-transitory processor readable storage medium for storing a computer program of instructions configured to be readable by at least one processor for instructing the at least one processor to execute a computer process for performing the method as recited in claim 1 .
12. The method of claim 1 , wherein the level of the increase is calculated to include a specified safety margin between administrative I/O requests and the maximum latency value.
13. An article of manufacture for storage network bandwidth management, the article of manufacture comprising:
at least one non-transitory processor readable medium; and
instructions stored on the at least one medium;
wherein the instructions are configured to be readable from the at least one medium by at least one processor and thereby cause the at least one processor to operate so as to:
sample application Input/Output (I/O) requests associated with a first unit of storage during a specified period of time, wherein the sampling is independent of volume management administrative I/O requests;
determine a maximum latency value based on the sampling of the application Input/Output (I/O) requests;
compare the maximum latency value with a current application I/O latency value; and
throttle the volume management administrative I/O requests in the event that the current application I/O latency value exceeds the maximum latency value, wherein the volume management administrative I/O requests are independent of the application Input/Output (I/O) requests; and
increasing the volume management administrative Input/Output (I/O) requests in the event the current application I/O latency value is a specified level below the maximum latency value, wherein the volume management administrative I/O requests are independent of the application Input/Output (I/O) requests and wherein a level of the increase is based upon an amount the current application I/O latency level is below the maximum latency value.
14. A system for storage network bandwidth management comprising:
one or more processors communicatively coupled to a network; wherein the one or more processors are configured to:
sample application Input/Output (I/O) requests associated with a first unit of storage during a specified period of time, wherein the sampling is independent of volume management administrative I/O requests;
determine a maximum latency value based on the sampling of the application Input/Output (I/O) requests;
compare the maximum latency value with a current application I/O latency value; and
throttle the volume management administrative I/O requests in the event that the current application I/O latency value exceeds the maximum latency value, wherein the volume management administrative I/O requests are independent of the application Input/Output (I/O) requests; and
increasing the volume management administrative Input/Output (I/O) requests in the event the current application I/O latency value is a specified level below the maximum latency value, wherein the volume management administrative I/O requests are independent of the application Input/Output (I/O) requests and wherein a level of the increase is based upon an amount the current application I/O latency level is below the maximum latency value.
15. The system of claim 14 , wherein the one or more processors are further configured to:
periodically sample the application Input/Output (I/O) requests to determine a new maximum latency value.
16. The system of claim 14 , wherein the first unit of storage comprises at least one of a volume, a disk, and a disk array.
17. The system of claim 14 , wherein the one or more processors are further configured to:
compare the current application I/O latency value with a maximum latency value associated with a second unit of storage, wherein throttling volume management administrative I/O requests comprises throttling the volume management administrative I/O requests in the event that the current application I/O latency value exceeds either the maximum latency value of the first unit of storage or the maximum latency value of the second unit of storage.
18. The system of claim 17 , wherein the comparison is based upon an administrative operation occurring which requires writing from the first unit of storage to the second unit of storage.
19. The system of claim 14 , wherein throttling volume management administrative I/O requests is implemented using a sliding window protocol.