Sensitivity matrix for system load indication and overload prevention

1. A method for performing input/output (IO) flow control in a storage system, the method comprising:

receiving an IO latency factor for each IO of a plurality of IOs in a workload, the IO latency factor specifying a priority level;

receiving a component latency factor, with respect to each of the Ios in the workload, for each component of a plurality of components in the storage system, the component latency factor indicating a degree to which the component is considered in assessing the workload;

processing the plurality of IOs in the workload, comprising:

applying, during processing of the workload, the IO latency factor and the component latency factor to each of the IOs in the workload; and

determining an effective average latency of the plurality of IOs in the workload as a function of the applied IO latency factors and the applied component latency factors; and

modifying the IO flow control of incoming IOs in the storage system as a function of the effective average latency.

2. The method of claim 1 , wherein modifying the IO flow control of the incoming IOs includes increasing an amount of the incoming IOs when the effective average latency is less than a threshold value.

3. The method of claim 1 , wherein applying the IO latency factor and the component latency factor to each of the corresponding IOs includes:

for each component of the plurality of components that processes the IO as determined by the component latency factor:

determining an amount of processing time attributed to the IO;

multiplying the amount of processing time by the component latency factor, yielding a component latency value;

summing component latency values for each of the components; and

multiplying the sum of the component latency values by the IO latency value, yielding an effective IO latency value attributed to the IO, the effective IO latency value specifying a load indicator.

4. The method of claim 3 , further comprising calculating a current average latency of the IOs processed in the workload comprising calculating effective IO latency values for each of the IOs processed in the workload, taking an average of the effective IO latency values.

5. The method of claim 4 , wherein determining the effective average latency includes trending the current average latency calculated for the IOs processed in the workload and a previous average latency calculated for IOs processed in a previous workload.

6. The method of claim 1 , wherein the IO latency factor is a value between 0 and 1, and the component latency factor is a value between 0 and 1.

7. The method of claim 1 , wherein the component latency factor measures an amount of time the IO is processed through a corresponding one of the components, and wherein the components subject to the component latency factor include at least one of:

synchronous replication remote data transfer functions; and

address lock resolution functions.

8. A system for performing input/output (IO) flow control in a storage system, the system comprising:

a memory comprising computer-executable instructions; and

a processor operable by a storage system, the processor executing the computer-executable instructions, the computer-executable instructions when executed by the processor cause the processor to perform operations comprising:

receiving an IO latency factor for each IO of a plurality of IOs in a workload, the IO latency factor specifying a priority level;

receiving a component latency factor, with respect to each of the IOs in the workload, for each component of a plurality of components in the storage system, the component latency factor indicating a degree to which the component is considered in assessing the workload;

processing the plurality of IOs in the workload, comprising:

applying, during processing of the workload, the IO latency factor and the component latency factor to each of the IOs in the workload; and

determining an effective average latency of the plurality of IOs in the workload as a function of the applied IO latency factors and the applied component latency factors; and

modifying the IO flow control of incoming IOs in the storage system as a function of the effective average latency.

9. The system of claim 8 , wherein modifying the IO flow control of the incoming IOs includes increasing an amount of the incoming IOs when the effective average latency is less than a threshold value.

10. The system of claim 8 , wherein applying the IO latency factor and the component latency factor to each of the corresponding IOs includes:

for each component of the plurality of components that processes the IO as determined by the component latency factor:

determining an amount of processing time attributed to the IO;

multiplying the amount of processing time by the component latency factor, yielding a component latency value;

summing component latency values for each of the components; and

multiplying the sum of the component latency values by the IO latency value, yielding an effective IO latency value attributed to the IO, the effective IO latency value specifying a load indicator.

11. The system of claim 10 , wherein the operations further comprise calculating a current average latency of the IOs processed in the workload comprising calculating effective IO latency values for each of the IOs processed in the workload, taking an average of the effective IO latency values.

12. The system of claim 11 , wherein determining the effective average latency includes trending the current average latency calculated for the IOs processed in the workload and a previous average latency calculated for IOs processed in a previous workload.

13. The system of claim 8 , wherein the IO latency factor is a value between 0 and 1, and the component latency factor is a value between 0 and 1.

14. The system of claim 8 , wherein the component latency factor measures an amount of time the IO is processed through a corresponding one of the components, and wherein the components subject to the component latency factor include at least one of:

synchronous replication remote data transfer functions; and

address lock resolution functions.

15. A computer program product performing input/output (IO) flow control in a storage system, the computer program product embodied on a non-transitory computer readable medium, the computer program product including instructions that, when executed by a computer, causes the computer to perform operations comprising:

receiving an IO latency factor for each IO of a plurality of IOs in a workload, the IO latency factor specifying a priority level;

receiving a component latency factor, with respect to each of the IOs in the workload, for each component of a plurality of components in the storage system, the component latency factor indicating a degree to which the component is considered in assessing the workload;

processing the plurality of IOs in the workload, comprising:

applying, during processing of the workload, the IO latency factor and the component latency factor to each of the IOs in the workload; and

determining an effective average latency of the plurality of IOs in the workload as a function of the applied IO latency factors and the applied component latency factors; and

modifying the IO flow control of incoming IOs in the storage system as a function of the effective average latency.

16. The computer program product of claim 15 , wherein modifying the IO flow control of the incoming IOs includes increasing an amount of the incoming IOs when the effective average latency is less than a threshold value.

17. The computer program product of claim 16 , wherein applying the IO latency factor and the component latency factor to each of the corresponding IOs includes:

for each component of the plurality of components that processes the IO as determined by the component latency factor:

determining an amount of processing time attributed to the IO;

multiplying the amount of processing time by the component latency factor, yielding a component latency value;

summing component latency values for each of the components; and

multiplying the sum of the component latency values by the IO latency value, yielding an effective IO latency value attributed to the IO, the effective IO latency value specifying a load indicator.

18. The computer program product of claim 17 , wherein the operations further comprise calculating a current average latency of the IOs processed in the workload comprising calculating effective IO latency values for each of the IOs processed in the workload, taking an average of the effective IO latency values.

19. The computer program product of claim 18 , wherein determining the effective average latency includes trending the current average latency calculated for the IOs processed in the workload and a previous average latency calculated for IOs processed in a previous workload.

20. The computer program product of claim 15 , wherein:

the IO latency factor is a value between 0 and 1, and the component latency factor is a value between 0 and 1;

the component latency factor measures an amount of time the IO is processed through a corresponding one of the components; and

the components subject to the component latency factor include at least one of:

synchronous replication remote data transfer functions; and

address lock resolution functions.