Browse Prior Art Database

Method and System for Implementing Hierarchical Scaling Pattern Technique in Cloud Computing Environment

IP.com Disclosure Number: IPCOM000236101D
Publication Date: 2014-Apr-04
Document File: 3 page(s) / 58K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system is disclosed for implementing hierarchical scaling pattern technique cloud computing environment.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Method and System for Implementing Hierarchical Scaling Pattern Technique in Cloud Computing Environment

Disclosed is a method and system for implementing hierarchical scaling pattern technique in cloud computing environment. The method and system enables a hierarchical approach to scaling patterns into arbitrarily large and geographically dispersed cloud computing environment. The hierarchical approach allows the virtual machine (VM) instances to be successfully deployed using placement policies selected by a user, such as, for example, high availability, energy conservation, etc.

VM instances are successfully deployed by scaling to an arbitrary number of

physical compute resources and deployed virtual server instances.

As shown in the figure, the hierarchical approach involves grouping of physical resources such as, for example, a rack, sub-domain, domain, etc. A rack contains

one or more compute nodes, storage, and data and storage switches and links. Whereas a site contains one or more racks and a region contains one or more sites. Different groupings could be made, and additional levels of hierarchy could be introduced by placing the engine at the domain level that selects the sub-domains into new virtual machine instances to scale the patterns.

Figure

In accordance with the method and system, while scaling a pattern, a pattern engine places new virtual machine instances onto a specific compute node in the cloud. This is accomplished by invoking a placement engine. At the rack level, the rack sees a set of physical computing nodes onto which it places each virtual machine instance, and the resources that the computing node provides. The rack-level placement is also aware of every virtual machine instance that is currently deployed to each computing node and the computing resources. Thereafter, while reporting

the resources available to the sub-domain placement engine, each rack summarizes its compute resources. Instead of reporting each compute node, an aggregate compute node for each host aggregate is reported that provides the total CPU, memory and storage available for that host aggregate. An aggregate virtual

machine instance is also reported that shows the total CPU, memory and storage

1


Page 02 of 3

that is reserved for virtual machine instances in that host aggregate. Further, while

reporting the physical resources available to the domain placement engine, each

sub-domain reports the host aggregates that may be deployed in that domain, along

with an aggregate amount of CPU, memory and storage for each rack's resources in that sub-domain. In cases where a host aggregate spans more than one rack each rack is reported as an independent compute resource.

The one or more virtual machine instances are added to an existing pattern while scaling a pattern. For each new virtual machine instances in a pattern, a set of instructions are specified. A set of placement...