Modeling the effects of switching data storage resources through data storage pool tier performance capacity and demand gap analysis

• US 10,216,458 B2
• Filed: 10/27/2017
• Issued: 02/26/2019
• Est. Priority Date: 12/19/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Sabloniere, Pierre
• Primary Examiner(s): Kim, Hong C
• Application Number:
  US15/795,316
• Publication Number:
  US 20180046412A1
• Time in Patent Office: 487 Days
• Field of Search: 711112, 711 4
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0647 Migration mechanisms
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

The assignment of data storage resources in a data storage arrangement having a plurality of pools, where each pool includes a plurality of data storage resources arranged in tiers, can be optimized. A gap analysis can be performed for each tier of each pool to determine that tier'"'"'s excess capacity or demand. The effect of switching data storage resources between tiers of different pools can be modeled to determine the effect of the switch on excess capacity and demand. An improved arrangement of data storage resources in the tiers can be determined that reduces excess capacity and demand, from the modeling of the effect of switching data storage resources between pools. At least one data storage resource can be switched from the tier of the first pool to the tier of the second pool to match the determined improved arrangement of data storage resources.

Claims

1. A computer-implemented method of optimizing assignment, without disruption to applications, of data storage resources in a data storage arrangement having a plurality of pools, each pool comprising a plurality of data storage resources arranged in tiers and controlled by a Redundant Array of Inexpensive Disks (RAID) controller, the method comprising:

performing, by a processor, a gap analysis for a plurality of tiers of a plurality of pools, the gap analysis including a comparison of performance capacity versus performance demand for each tier of the plurality of tiers of the plurality of pools, the gap analysis also determining an extent to which tiers of pools have, relative to a standard data storage response time performance metric, excess performance capacity and excess performance demand, measured over a set period of time;

modeling, by a processor, an effect of switching data storage resources from a first pool tier to a second pool tier in terms of the effect of the switch on excess performance capacity and excess performance demand;

determining, by a processor, an improved arrangement of data storage resources in the first pool tier and the second pool tier that reduces excess performance capacity and excess performance demand, based on the modeling of the effect of switching data storage resources between pools; and

switching at least one data storage resource from the first pool tier to the second pool tier based on the modeling.

2. The method of claim 1, wherein the switching of at least one data storage resource from the first pool tier to the second pool tier is performed if a gain for reducing excess performance capacity and excess performance demand is above a predetermined threshold.

3. The method of claim 1, and further comprising, prior to the switching of at least one data storage resource from the first pool tier to the second pool tier, delaying for a predetermined length of time and rerunning the gap analysis, the modeling and the determining.

4. The method of claim 1, wherein the determining of the improved arrangement of data storage resources in the first pool tiers and the second pool tiers for reducing excess performance capacity and excess performance demand comprises matching excess performance capacity to excess performance demand in corresponding tiers in different pools to reduce overall excess performance capacity and excess performance demand.