Granted Patent B2

US 10,095,422 · App. 15/337,667 · Granted Oct 9, 2018

Abstract

Claims (51)

1. A computer-implemented method for allocating input/output bandwidth in storage systems, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising:

allocating, in accordance with a pre-defined service-level agreement, an assigned amount of input/output bandwidth to an application that performs actions on a storage system;

receiving a request from the application to perform an input/output operation on the storage system;

identifying an amount of latency associated with fulfilling the input/output operation, wherein the amount of latency represents a severity of impact on the ability of the input/output system to fulfill other requests to perform input/output operations on the storage system;

calculating an amount of input/output bandwidth consumed by the input/output operation that is proportional to the latency associated with fulfilling the input/output operation instead of the size of the input/output operation;

deducting the amount of input/output bandwidth consumed by the input/output operation from the assigned amount of input/output bandwidth allocated to the application; and

allocating bandwidth to future input/output requests from the application from the remaining amount of input/output bandwidth allocated to the application.

2. The method of claim 1 , wherein identifying the amount of latency associated with fulfilling the input/output operation comprises measuring an amount of time taken to complete the input/output operation.

3. The method of claim 1 , wherein identifying the amount of latency associated with fulfilling the input/output operation comprises calculating an estimated amount of latency that will be induced by the input/output operation based at least in part on whether the input/output operation comprises a plurality of sequential input/output operations.

4. The method of claim 3 , wherein calculating the estimated amount of latency comprises calculating the estimated amount of latency based on previous observations of input/output operations that are similar to the input/output operation.

5. The method of claim 1 , wherein receiving the request from the application comprises receiving a request to perform a plurality of input/output operations.

6. The method of claim 5 , wherein receiving the request from the application comprises treating the plurality of input/output operations as a single input/output operation.

7. The method of claim 6 , wherein treating the plurality of input/output operations as a single input/output operation comprises:

calculating the total amount of input/output bandwidth consumed by the plurality of input/output operations; and

deducting the total amount of input/output bandwidth consumed by the plurality of input/output operations.

8. The method of claim 1 , wherein the request from the application comprises a request to perform at least one of:

a backup operation;

a restore operation;

a recovery operation; and

a replication operation.

9. The method of claim 1 , wherein allocating the assigned amount of input/output bandwidth to the application comprises allocating the assigned amount of input/output bandwidth based on both the number of applications that request input/output bandwidth from the storage system and the capacity of the storage system to perform input/output operations.

10. The method of claim 1 , wherein allocating bandwidth to future input/output requests from the application from the remaining amount of input/output bandwidth allocated to the application comprises preventing the application from performing at least one future input/output operation on the storage system in response to determining that the application has an insufficient amount of input/output bandwidth remaining to perform the future input/output operation.

11. The method of claim 1 , wherein:

allocating the assigned amount of input/output bandwidth to the application comprises allocating an assigned number of credits to the application;

calculating the amount of input/output bandwidth consumed by the input/output operation comprises converting the latency associated with fulfilling the input/output operation into a number of consumed credits that are consumed by the input/output operation; and

deducting the amount of input/output bandwidth consumed by the input/output operation comprises subtracting the number of consumed credits from the assigned number of credits.

12. The method of claim 11 , wherein converting the latency associated with fulfilling the input/output operation into a number of consumed credits comprises converting the latency associated with fulfilling the input/output operation based on a reference latency that represents one credit.

13. A system for allocating input/output bandwidth in storage systems, the system comprising:

an allocation module, stored in memory, that allocates, in accordance with a pre-defined service-level agreement, an assigned amount of input/output bandwidth to an application that performs actions on a storage system;

a receiving module, stored in memory, that receives a request from the application to perform an input/output operation on the storage system;

an identification module, stored in memory, that identifies an amount of latency associated with fulfilling the input/output operation, wherein the amount of latency represents a severity of impact on the ability of the input/output system to fulfill other requests to perform input/output operations on the storage system;

a calculation module, stored in memory, that calculates an amount of input/output bandwidth consumed by the input/output operation that is proportional to the latency associated with fulfilling the input/output operation instead of the size of the input/output operation;

a deduction module, stored in memory, that deducts the amount of input/output bandwidth consumed by the input/output operation from the assigned amount of input/output bandwidth allocated to the application; and

at least one physical processor configured to execute the allocation module, the receiving module, the identification module, the calculation module, and the deduction module.

14. The system of claim 13 , wherein the allocation module allocates bandwidth to future input/output requests from the application from the remaining amount of input/output bandwidth allocated to the application.

15. The system of claim 13 , wherein the identification module identifies the amount of latency associated with fulfilling the input/output operation by measuring an amount of time taken to complete the input/output operation.

16. The system of claim 13 , wherein the identification module identifies the amount of latency associated with fulfilling the input/output operation by calculating an estimated amount of latency that will be induced by the input/output operation based at least in part on whether the input/output operation comprises a plurality of sequential input/output operations.

17. The system of claim 16 , wherein the identification module calculates the estimated amount of latency by calculating the estimated amount of latency based on previous observations of input/output operations that are similar to the input/output operation.

18. The system of claim 13 , wherein the receiving module receives the request from the application by receiving a request to perform a plurality of input/output operations.

19. The system of claim 13 , wherein the request from the application comprises a request to perform at least one of:

a backup operation;

a restore operation;

a recovery operation; and

a replication operation.

20. A non-transitory computer-readable medium comprising one or more computer-readable instructions that, when executed by at least one processor of a computing device, cause the computing device to:

allocate, in accordance with a pre-defined service-level agreement, an assigned amount of input/output bandwidth to an application that performs actions on a storage system;

receive a request from the application to perform an input/output operation on the storage system;

identify an amount of latency associated with fulfilling the input/output operation, wherein the amount of latency represents a severity of impact on the ability of the input/output system to fulfill other requests to perform input/output operations on the storage system;

calculate an amount of input/output bandwidth consumed by the input/output operation that is proportional to the latency associated with fulfilling the input/output operation instead of the size of the input/output operation;

deduct the amount of input/output bandwidth consumed by the input/output operation from the assigned amount of input/output bandwidth allocated to the application; and

allocate bandwidth to future input/output requests from the application from the remaining amount of input/output bandwidth allocated to the application.

Assignments (13)