Method, system and computer readable medium for controlling performance of storage pools

Patent metadata

Field	Value
Publication number	US-9940033-B1
Application number	US-201514984296-A
Country	US
Kind code	B1
Filing date	Dec 30, 2015
Priority date	Dec 30, 2015
Publication date	Apr 10, 2018
Grant date	Apr 10, 2018

Abstract

Described are techniques for controlling performance. A forecast error Et for time period “t” denoting a difference between a predicted performance value Ft of a metric for time period “t” and an actual performance value At of the metric for time period “t” is determined. A tracking signal Tt for time period “t” is determined in accordance with “N” forecast errors. It is determined, in accordance with the tracking signal Tt for time period “t”, whether to modify a current value for alpha denoting a weight used in calculating predicted performance values for the metric. If it is determined to modify the current value for alpha, the current value for alpha is set to a revised value in accordance with the tracking signal Tt for time period “t”. A predicted performance value F t+1 of the metric for time period “t+1” is determined in accordance with the current value for alpha.

First claim

What is claimed is:

1. A method of controlling performance of storage pools comprising:
determining, using a processor, a forecast error Et for time period “t” denoting a difference between a predicted performance value Ft of a metric for time period “t” and an actual performance value At of the metric for time period “t”;
determining, using a processor, a tracking signal Tt for time period “t” in accordance with “N” forecast errors including the forecast error Et for time period “t”, wherein the tracking signal Tt for time period “t” is a ratio of a first value with respect to a second value, wherein the first value is a sum of the “N” forecast errors, and the second value is a mean absolute error or deviation determined with respect to the “N” forecast errors;
determining, using a processor and in accordance with the tracking signal Tt for time period “t”, whether to modify a current value for alpha denoting a weight used in calculating predicted performance values for the metric;
if it is determined to modify the current value for alpha, setting, using a processor, the current value for said alpha to a revised value in accordance with the tracking signal Tt for time period “t”;
determining, using a processor, a predicted performance value F t+1 of the metric for time period “t+1” in accordance with the current value for said alpha, wherein Et, Ft, Tt and F t+1 are determined with respect to a first storage pool of one or more storage devices;
determining, using a processor and in accordance with the predicted performance value F t+1 for the metric for time period “t+1”, whether there is a performance violation for the first storage pool; and
responsive to determining there is a performance violation, performing, using a processor, one or more data movements to reduce I/O workload of the first storage pool;
wherein the one or more data movements include moving a data portion from the first storage pool to a target storage pool, the data portion having a first I/O workload;
wherein the target storage pool has available capacity and wherein modeling determines that the target storage pool accepting the first I/O workload does not result in the target storage pool having a performance violation.

2. The method of claim 1 , wherein the target storage pool has a performance ranking that is equal to or higher than a performance ranking of the first storage pool.

3. The method of claim 1 , wherein the metric is average I/O response time for the first storage pool.

4. The method of claim 1 , wherein it is determined to modify the current value for said alpha by increasing the current value for said alpha.

5. The method of claim 1 , wherein it is determined to modify the current value for said alpha by decreasing the current value for said alpha.

6. The method of claim 1 , wherein the second value is a result of dividing a sum of absolute values of the “N” forecast errors by “N”.

7. The method of claim 1 , further comprising:
determining whether the tracking signal Tt for time period “t” is within a first defined subrange of tracking signal values;
determining whether two tracking signals of two consecutive time periods “t−1” and “t−2” immediately prior to time period “t” are within the first defined subrange of tracking signal values;
if it is determined that the tracking signal Tt for time period “t” and two tracking signals of two consecutive time periods “t−1” and “t−2” immediately prior to time period “t” are within the first defined subrange of tracking signal values, modifying the current value of said alpha in accordance with the first defined subrange.

8. The method of claim 7 , wherein the first defined subrange denotes tracking signal values for which the current value for said alpha should be decreased and wherein the current value of said alpha is accordingly decreased.

9. The method of claim 7 , wherein the first defined subrange denotes tracking signal values for which the current value for said alpha should be increased and wherein the current value of said alpha is accordingly increased.

10. The method of claim 1 , wherein said alpha is used in determining a weighting of the actual performance value At of the metric for time period “t” and a weighting of the predicted performance value Ft of the metric for time period “t” in connection with determining the predicted performance value F t+1 of the metric for time period “t+1”.

11. The method of claim 10 , wherein the current value for said alpha denotes the weighting of the actual performance value At of the metric for time period “t” in connection with determining the predicted performance value F t+1 of the metric for time period “t+1”.

12. The method of claim 11 , wherein one, less the current value for said alpha, denotes the weighting of the predicted performance value Ft of the metric for time period “t” in connection with determining the predicted performance value F t+1 of the metric for time period “t+1”.

13. A system comprising:
a processor; and
a memory comprising code stored thereon that, when executed by the processor, performs a method of controlling performance of storage pools comprising:
determining a forecast error Et for time period “t” denoting a difference between a predicted performance value Ft of a metric for time period “t” and an actual performance value At of the metric for time period “t”;
determining a tracking signal Tt for time period “t” in accordance with “N” forecast errors including the forecast error Et for time period “t”, wherein the tracking signal Tt for time period “t” is a ratio of a first value with respect to a second value, wherein the first value is a sum of the “N” forecast errors, and the second value is a mean absolute error or deviation determined with respect to the “N” forecast errors;
determining, in accordance with the tracking signal Tt for time period “t”, whether to modify a current value for alpha denoting a weight used in calculating predicted performance values for the metric;
if it is determined to modify the current value for alpha, setting the current value for said alpha to a revised value in accordance with the tracking signal Tt for time period “t”;
determining a predicted performance value F t+1 of the metric for time period “t+1” in accordance with the current value for said alpha, wherein Et, Ft, Tt and F t+1 are determined with respect to a first storage pool of one or more storage devices;
determining, using a processor and in accordance with the predicted performance value F t+1 for the metric for time period “t+1”, whether there is a performance violation for the first storage pool; and
responsive to determining there is a performance violation, performing, using a processor one or more data movements to reduce I/O workload of the first storage pool;

wherein the one or more data movements include moving a data portion from the first storage pool to a target storage pool, the data portion having a first I/O workload;
wherein the target storage pool has available capacity and wherein modeling determines that the target storage pool accepting the first I/O workload does not result in the target storage pool having a performance violation.

14. A non-transitory computer readable medium comprising code stored thereon that, when executed, performs a method of controlling performance of storage pools comprising:
determining a forecast error Et for time period “t” denoting a difference between a predicted performance value Ft of a metric for time period “t” and an actual performance value At of the metric for time period “t”;
determining a tracking signal Tt for time period “t” in accord

Assignees

• Emc Corp
• Emc Ip Holding Co Llc

Inventors

• Alshawabkeh Malak
• Martin Owen

Classifications

• G06F3/0685

  Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays · CPC title

• G06F11/3452

  Performance evaluation by statistical analysis · CPC title

• G06F11/3034

  where the computing system component is a storage system, e.g. DASD based or network based (digital input from or digital output to record carriers G06F3/06; digital recording or reproducing G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097) · CPC title

• G06F11/3409

  for performance assessment · CPC title

• G06F11/3433 Primary

  for load management (allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029) · CPC title