Allocation and balancing of storage resources

Patent metadata

Field	Value
Publication number	US-10528266-B2
Application number	US-201615364029-A
Country	US
Kind code	B2
Filing date	Nov 29, 2016
Priority date	Feb 3, 2012
Publication date	Jan 7, 2020
Grant date	Jan 7, 2020

Abstract

A method and technique for allocation and balancing of storage resources includes monitoring, for each of a plurality of storage controllers, an input/output (I/O) latency value based on an I/O latency associated with each storage volume controlled by a respective storage controller. A network bandwidth utilization value and a network latency value corresponding to each storage controller is also monitored. An I/O latency value threshold, a network bandwidth utilization threshold, and a network latency value threshold is determined. Responsive to at least one of the I/O latency value, the network bandwidth utilization value and the network latency value for at least one storage controller exceeding a respective threshold, storage volume distribution among the storage controllers is rebalanced.

First claim

What is claimed is:

1. A method, comprising:
determining an average storage volume input/output (I/O) latency for each of a plurality of storage volumes;
calculating an I/O latency value for each of a plurality of storage controllers based on the average storage volume I/O latency of each storage volume controlled by the respective storage controller;
monitoring, for each of the plurality of storage controllers the respective I/O;
monitoring a network bandwidth utilization value and a network latency value corresponding to each storage controller;
determining an I/O latency value threshold, a network bandwidth utilization threshold, and a network latency value threshold; and
responsive to a change in at least one of the I/O latency value, the network bandwidth utilization value and the network latency value for at least one storage controller exceeding a respective threshold, rebalancing storage volume distribution among the storage controllers.

2. The method of claim 1 , further comprising weighting each storage volume based on a quantity of data being processed by the respective storage volume.

3. The method of claim 1 , wherein calculating the I/O latency value for each storage controller includes calculating an average I/O latency value for each storage controller.

4. The method of claim 3 , wherein calculating the average I/O latency for each storage controller comprises:
calculating a weighted average I/O latency value for each storage volume controlled by a respective storage controller;
summing the weighted I/O latency values of the storage volumes controlled by the respective storage controller; and
dividing the summed I/O latency values by a quantity of the storage volumes controlled by the respective storage controller.

5. The method of claim 4 , wherein calculating the weighted average I/O latency value for each storage volume comprises calculating the weighted average I/O latency value based on weighting write I/O operations greater than read I/O operations.

6. The method of claim 1 , further comprising:
determining a type of application utilizing one or more of the storage volumes;
determining a priority level of the application; and
rebalancing the storage volume distribution among the storage controllers based on the type of the application and the priority level of the application.

7. The method of claim 1 , further comprising:
analyzing the I/O latency value, the network bandwidth utilization value and the network latency value for the storage controllers;
determining whether moving one or more storage volumes will lower at least one of the I/O latency value, the network bandwidth utilization value and the network latency value for at least one storage controller; and
responsive to determining that moving one or more storage volumes will lower at least one of the I/O latency value, the network bandwidth utilization value and the network latency value for at least one storage controller, migrating at least one storage volume from one storage controller to another storage controller.

8. A system, comprising:
a plurality of storage controllers each controlling one or more storage volumes; and
a processor unit operable to execute a management application, the management application configured to:
determine an average storage volume input/output (I/O) latency for each of the one or more storage volumes;
calculate an I/O latency value for each of the plurality of storage controllers based on the average storage volume I/O latency of each storage volume controlled by the respective storage controller;
monitor, for each of the plurality of storage controllers, the respective I/O latency value;
monitor a network bandwidth utilization value and a network latency value corresponding to each storage controller;
determine an I/O latency value threshold, a network bandwidth utilization threshold, and a network latency value threshold; and
responsive to a change in at least one of the I/O latency value, the network bandwidth utilization value and the network latency value for at least one storage controller exceeding a respective threshold, rebalance storage volume distribution among the storage controllers.

9. The system of claim 8 , wherein the management application is configured to weight each storage volume based on a quantity of data being processed by the respective storage volume.

10. The system of claim 8 , wherein the management application is configured to calculate an average I/O latency value for each storage controller.

11. The system of claim 10 , wherein the management application is configured to calculate the average I/O latency for each storage controller by:
calculating a weighted average I/O latency value for each storage volume controlled by a respective storage controller;
summing the weighted I/O latency values of the storage volumes controlled by the respective storage controller; and
dividing the summed I/O latency values by a quantity of the storage volumes controlled by the respective storage controller.

12. The system of claim 8 , wherein the I/O latency value for each storage controller includes a weighted average I/O latency value, wherein the weighting is based on a rate of I/O operations.

13. The system of claim 8 , wherein the management application is configured to:
determine a type of application utilizing one or more of the storage volumes;
determine a priority level of the application; and
rebalance the storage volume distribution among the storage controllers based on the type of the application and the priority level of the application.

14. The system of claim 8 , wherein the management application is configured to:
analyze the I/O latency value, the network bandwidth utilization value and the network latency value for the storage controllers;
determine whether moving one or more storage volumes will lower at least one of the I/O latency value, the network bandwidth utilization value and the network latency value for at least one storage controller; and
responsive to determining that moving one or more storage volumes will lower at least one of the I/O latency value, the network bandwidth utilization value and the network latency value for at least one storage controller, migrate at least one storage volume from one storage controller to another storage controller.

15. A computer program product for storage management, the computer program product comprising:
a non-transitory computer readable medium having computer readable program code embodied therewith, the computer readable program code comprising computer readable program code configured to:
determine an average storage volume input/output (I/O) latency for each of a plurality of storage volumes;
calculate an I/O latency value for each of a plurality of storage controllers based on the average storage volume I/O latency of each storage volume controlled by the respective storage controller;
monitor, for each of the plurality of storage controllers the respective I/O;
monitor a network bandwidth utilization value and a network latency value corresponding to each storage controller;
determine an I/O latency value threshold, a network bandwidth utilization threshold, and a network latency value threshold; and
responsive to a change in at least one of the I/O latency value, the network bandwidth utilization value and the network latency value for at least one storage controller exceeding a respective threshold, rebalance storage volume distribution among the storage controllers.

16. The computer program product of claim 15 , wherein the computer

Assignees

• IBM

Inventors

• Ashok Rohith K
• Brabson Roy F
• Chan Wei L

Classifications

• G06F3/0613

  in relation to throughput · CPC title

• G06F9/5011

  the resources being hardware resources other than CPUs, Servers and Terminals · CPC title

• G06F3/061

  Improving I/O performance · CPC title

• G06F3/0631 Primary

  by allocating resources to storage systems · CPC title

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title