Browse Prior Art Database

Method and System for Provisioning Cloud Virtual Machines to Ensure Mapping to Independent Hardware Nodes

IP.com Disclosure Number: IPCOM000200546D
Publication Date: 2010-Oct-18
Document File: 3 page(s) / 25K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method and algorithm to enforce the selection of independent nodes for a set of virtual machines in a Cloud environment. In the event of a hardware failure, the component based High Availability Disaster Recovery (HADR) architecture running in the Cloud continues to operate as designed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 54% of the total text.

Page 01 of 3

Method and System for Provisioning Cloud Virtual Machines to Ensure Mapping to

Independent Hardware Nodes


Cloud environments are emerging as popular solutions for organizations needing to

host complete operational production infrastructures in one location. This may include

all components of a tiered application -- web, logic, and database layers -- entirely

hosted in the Cloud. Independent Software Vendors (ISV) are providing component

level High Availability Disaster Recovery (HADR) capabilities. For example, a DB2

cluster operating across multiple virtual machines (VM) in a Cloud provides DB2

component level HADR for two or more DB2 VMs running in a cluster on the Cloud. There is a risk however, if during provisioning all the VMs are hosted on the same

hardware node, and that node fails, the entire set of components fails, and the ISV

provided HA cluster provides no value.

This invention provides a method and algorithm to enforce the selection of independent

nodes for a set of virtual machines in a Cloud environment. In the event of a hardware

failure, the component based HADR architecture running in the Cloud continues to

operate as designed.

The invention eliminates the possibility of provisioning virtual machine images on the

same hardware node, at least in accordance with available hardware capacity in a

Cloud environment. The algorithm is shown in two modes:

• In initial batch provisioning

• In a dynamic management scenario with the uniqueness/independence of a new

target Cloud hardware node during a virtual machine migration.

Ultimately this algorithm can be integrated into a provisioning component or a

monitoring and management runtime system for Cloud env...