Methods and systems for managing service level objectives in a networked storage environment

1. A method, comprising:

setting, by a processor, an attribute for a first workload for enabling dynamic service level objective (SLO) adjustment in a networked storage system;

determining by the processor, a demand pattern for the first workload that is assigned a SLO for using a resource of the networked storage system, the SLO defined by an allotted performance parameter for storing and retrieving data from the network storage system;

utilizing, by the processor, the demand pattern to identify a first duration when SLO allotment for the first workload is underutilized, and a second duration when the SLO allotment is being utilized;

based on the set attribute, dynamically adjusting the SLO allotment by the processor for the first duration by modifying a parameter associated with the allotted performance parameter, while maintaining the SLO allotment for the second duration;

upon modifying the parameter, determining by the processor, additional performance capacity of the resource available for re-allocation; the additional available performance capacity of the resource is based on a relationship between latency and utilization of the resource, the relationship selected from latency and utilization relationships generated by a model based technique and an observation based technique; and

dynamically allocating the additional available performance capacity by the processor for an identified second workload that needs an increase in SLO allotment for a certain duration.

2. The method of claim 1 , further comprising: dynamically allocating the additional performance capacity of the resource by the processor for provisioning a new workload in the networked storage system.

3. The method of claim 1 , further comprising: utilizing a number of input/output operations per second (IOPS) as the modified parameter, the number of IOPS budgeted for the resource for storing and retrieving data.

4. The method of claim 1 , further comprising: utilizing a target latency for the first workload as the modified parameter, the target latency associated with a throughput, where the throughput is the allotted performance parameter.

5. The method of claim 1 , further comprising: for the model based technique, using by the processor, a queuing model with inter-arrival time and service time for processing workload requests to generate a relationship between latency and utilization.

6. The method of claim 1 , further comprising: for the observation based technique, using by the processor measured latency and utilization of the resource to generate a relationship between latency and utilization.

7. The method of claim 1 , wherein the allotted performance parameter is a number of input/output operations per second (IOPS) allotted to the first workload for storing and retrieving data.

8. A non-transitory machine readable storage medium having stored thereon instructions for performing a method, comprising machine executable code which when executed by at least one machine, causes the machine to:

determine a demand pattern for a first workload that is assigned a service level objective (SLO) for using a resource of a networked storage system, the SLO defined by an allotted performance parameter for storing and retrieving data from the network storage system;

utilize the demand pattern to identify a first duration when SLO allotment for the first workload is underutilized and a second duration when the SLO allotment is being utilized;

based on a set attribute of the first workload, dynamically adjust the SLO allotment for the first duration, by modifying a parameter associated with the allotted performance parameter, while maintaining the SLO allotment for the second duration;

upon modifying the parameter, determine additional performance capacity of the resource available for re-allocation; the additional available performance capacity of the resource is based on a relationship between latency and utilization of the resource, the relationship selected from latency and utilization relationships generated by a model based technique and an observation based technique; and

dynamically allocate the additional available performance capacity for an identified second workload that needs an increase in SLO allotment for a certain duration.

9. The storage medium of claim 8 , wherein the machine executable code further causes the machine to: dynamically allocate the additional performance capacity of the resource for provisioning a new workload in the networked storage system.

10. The storage medium of claim 8 , wherein the machine executable code further causes the machine to: utilize a number of input/output operations per second (IOPS) as the modified parameter, the number of IOPS budgeted for the resource for storing and retrieving data.

11. The storage medium of claim 8 , wherein the machine executable code further causes the machine to: utilize a target latency for the first workload as the modified parameter, the target latency associated with a throughput, where the throughput is the allotted performance parameter.

12. The storage medium of claim 8 , wherein the machine executable code further causes the machine to: use a queuing model with inter-arrival time and service time for processing workload requests to generate a relationship between latency and utilization for the model based technique.

13. The storage medium of claim 8 , wherein the machine executable code further causes the machine to: use measured latency and utilization of the resource to generate a relationship between latency and utilization for the observation based technique.

14. The storage medium of claim 8 , wherein the allotted performance parameter is a number of input/output operations per second (IOPS) allotted to the first workload for storing and retrieving data.

15. A system comprising:

a memory containing machine readable medium comprising machine executable code having stored thereon instructions; and a processor coupled to the memory, the processor configured to execute the machine executable code to:

determine a demand pattern for a first workload that is assigned a service level objective (SLO) for using a resource of a networked storage system, the SLO defined by an allotted performance parameter for storing and retrieving data from the network storage system;

utilize the demand pattern to identify a first duration when SLO allotment for the first workload is underutilized, and a second duration when the SLO allotment is being utilized;

based on a set attribute for the first workload, dynamically adjust the SLO allotment for the first duration by modifying a parameter associated with the allotted performance parameter, while maintaining the SLO allotment for the second duration;

upon modifying the parameter, determine additional performance capacity of the resource available for re-allocation; the additional available performance capacity of the resource is based on a relationship between latency and utilization of the resource, the relationship selected from latency and utilization relationships generated by a model based technique and an observation based technique; and

dynamically allocate the additional available performance capacity for an identified second workload that needs an increase in SLO allotment for a certain duration.

16. The system of claim 8 , wherein the processor further executes the machine executable code to: dynamically allocate the additional performance capacity of the resource for provisioning a new workload in the networked storage system.

17. The system of claim 15 , wherein the processor further executes the machine executable code to: utilize a number of input/output operations per second (IOPS) as the modified parameter, the number of IOPS budgeted for the resource for storing and retrieving data.

18. The system of claim 15 , wherein the processor further executes the machine executable code to: use a queuing model with inter-arrival time and service time for processing workload requests to generate a relationship between latency and utilization for the model based technique.

19. The system of claim 15 , wherein the processor further executes the machine executable code to: use measured latency and utilization of the resource to generate a relationship between latency and utilization for the observation based technique.

20. The system of claim 15 , wherein the allotted performance parameter is a number of input/output operations per second (IOPS) allotted to the first workload for storing and retrieving data.