Selecting resources for automatic modeling using forecast thresholds

Patent metadata

Field	Value
Publication number	US-10142179-B2
Application number	US-201514645647-A
Country	US
Kind code	B2
Filing date	Mar 12, 2015
Priority date	Mar 12, 2015
Publication date	Nov 27, 2018
Grant date	Nov 27, 2018

Abstract

A method includes retrieving capacity utilization data for a plurality of resources and applying a linear regression analysis on the capacity utilization data. The method further includes projecting, using a processor, the capacity utilization data through a future time based on results of the linear regression analysis. The method additionally includes determining a deviation from a predetermined threshold range in the projected capacity utilization data for a first resource, and, in response to determining the deviation, determining, for each of a plurality of resource configurations, future capacity utilization of the first resource based on a non-linear capacity consumption model corresponding to the resource configuration. The method also includes applying a selected resource configuration from the plurality of resource configurations to the first resource to prevent the first resource from deviating from the predetermined threshold range.

First claim

What is claimed is:

1. A method, comprising:
retrieving capacity utilization data for a plurality of resources;
applying a linear regression analysis on the capacity utilization data;
projecting, using a processor, the capacity utilization data through a future time based on results of the linear regression analysis;
determining a deviation from a predetermined threshold range in the projected capacity utilization data for a first resource on a host, wherein the first resource is a virtual processor;
in response to determining the deviation, automatically selecting a plurality of alternative resource configurations for the plurality of resources based on predefined resource templates that specify resource options available for the alternative resource configurations, wherein each of the alternative resource configurations is selected based on whether it allocates additional resources to the first resource;
in further response to determining the deviation, automatically determining, for each of the plurality of alternative resource configurations, future capacity utilization of the first resource based on a non-linear capacity consumption model corresponding to the resource configuration, wherein the non-linear capacity consumption model comprises a plurality of resource scores that each characterize a capacity of each resource in the resource configuration to service additional workloads in view of scalability characteristics of a virtual memory and virtual storage I/O on the host in the non-linear capacity consumption model, and wherein the future capacity utilization of the first resource is dependent upon the scalability characteristics of the host virtual memory and virtual storage I/O; and
applying a selected resource configuration from the plurality of alternative resource configurations to the first resource to prevent the first resource from deviating from the predetermined threshold range.

2. The method of claim 1 , wherein the first resource is a virtual resource associated with a particular service, and wherein at least one of the plurality of alternative resource configurations comprises configuration information regarding other virtual resources associated with the particular service.

3. The method of claim 1 , further comprising:
receiving a growth rate associated with a workload of the first resource, and wherein determining the future capacity utilization of the first resource further comprises modifying the workload of the first resource by the growth rate.

4. The method of claim 1 , wherein the first resource is a server, and wherein the selected resource configuration comprises a virtual resource placement map for the server.

5. The method of claim 1 , wherein the selected resource configuration comprises a consolidation of physical servers associated with the plurality of resources.

6. The method of claim 1 , further comprising:
applying a set of related configuration changes to the plurality of resources based on the selected resource configuration.

7. The method of claim 1 , wherein projecting the capacity utilization data comprises projecting CPU capacity utilization and memory capacity utilization for each of the plurality of resources.

8. A computer configured to access a storage device, the computer comprising:
a processor; and
a non-transitory, computer-readable storage medium storing computer-readable instructions that when executed by the processor cause the computer to perform:
retrieving capacity utilization data for a plurality of resources;
applying a linear regression analysis on the capacity utilization data;
projecting, using a processor, the capacity utilization data through a future time based on results of the linear regression analysis;
determining a deviation from a predetermined threshold range in the projected capacity utilization data for a first resource on a host, wherein the first resource is a virtual processor;
in response to determining the deviation, automatically selecting a plurality of alternative resource configurations for the plurality of resources based on predefined resource templates that specify resource options available for the alternative resource configurations, wherein each of the alternative resource configurations is selected based on whether it allocates additional resources to the first resource;
in further response to determining the deviation, automatically determining, for each of the plurality of alternative resource configurations, future capacity utilization of the first resource based on a non-linear capacity consumption model corresponding to the resource configuration, wherein the non-linear capacity consumption model comprises a plurality of resource scores that each characterize a capacity of each resource in the resource configuration to service additional workloads in view of scalability characteristics of a virtual memory and virtual storage I/O on the host in the non-linear capacity consumption model, and wherein the future capacity utilization of the first resource is dependent upon the scalability characteristics of the host virtual memory and virtual storage I/O; and
applying a selected resource configuration from the plurality of alternative resource configurations to the first resource to prevent the first resource from deviating from the predetermined threshold range.

9. The computer of claim 8 , wherein the first resource is a virtual resource associated with a particular service, and wherein at least one of the plurality of alternative resource configurations comprises configuration information regarding other virtual resources associated with the particular service.

10. The computer of claim 8 , wherein the computer-readable instructions further cause the computer to perform:
receiving a growth rate associated with a workload of the first resource, and wherein determining the future capacity utilization of the first resource further comprises modifying the workload of the first resource by the growth rate.

11. The computer of claim 8 , wherein the first resource is a server, and wherein the selected resource configuration comprises a virtual resource placement map for the server.

12. The computer of claim 8 , wherein the selected resource configuration comprises a consolidation of physical servers associated with the plurality of resources.

13. The computer of claim 8 , wherein the computer-readable instructions further cause the computer to perform:
applying a set of related configuration changes to the plurality of resources based on the selected resource configuration.

14. The computer of claim 8 , wherein projecting the capacity utilization data comprises projecting CPU capacity utilization and memory capacity utilization for each of the plurality of resources.

15. A computer program product comprising:
a non-transitory computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code comprising:
computer-readable program code configured to retrieve capacity utilization data for a plurality of resources;
computer-readable program code configured to apply a linear regression analysis on the capacity utilization data;
computer-readable program code configured to project, using a processor, the capacity utilization data through a future time based on results of the linear regression analysis;
computer-readable program code configured to determine a deviation from a predetermined threshold range in the projected capacity utilization data for a first resource on a host, wherein the first resource is a virtual processor;
computer-readable prog

Assignees

• Ashby Jr John Wiley
• Varadaraju Balaji
• Ca Inc

Inventors

• Ashby Jr John Wiley
• Varadaraju Balaji

Classifications

• H04L41/145

  involving simulating, designing, planning or modelling of a network · CPC title

• H04L41/0823 Primary

  characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability (for optimising operational conditions of wireless networks H04W24/02) · CPC title

• H04L43/0817

  by checking functioning · CPC title

• H04L43/0876

  Network utilisation, e.g. volume of load or congestion level · CPC title

• H04L41/22

  comprising specially adapted graphical user interfaces [GUI] · CPC title