Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

A cloud platform-layer system and method to orchestrate ordering and rolling during VMs maintenance with optimization execution plan

IP.com Disclosure Number: IPCOM000242011D
Publication Date: 2015-Jun-14
Document File: 4 page(s) / 189K

Publishing Venue

The IP.com Prior Art Database

Abstract

A cloud platform layer system and method to orchestrate ordering and rolling during VMs maintenance with optimization execution plan is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 4

A cloud platform-layer system and method to orchestrate ordering and rolling during VMs maintenance with optimization execution plan

Disclosed is a cloud platform layer system and method to orchestrate ordering and rolling during VMs maintenance with optimization execution plan.

Platform as a Service (PaaS) is the delivery of a cloud platform on which developers can create and deploy middleware, applications and cloud services. PaaS has advantages including low cost of entry, elimination of infrastructure management, and ease of workload deployment and support. There are many PaaS providers which have different capabilities to develop and deploy applications. Pattern definition may include virtual systems and virtual applications. Patterns enable users to repetitively deploy middleware stacks across virtual machines with controlled sequences by the pattern creator. At least two types of roles are involved with the pattern creation process. One role is the pattern creator who uses the pattern builder to create the pattern. The other role is a content provider, who produces the content used to create the pattern. Also, there may be a pattern deplorer role deploying a previously created pattern into the cloud infrastructure. In the Cloud platform, there are many deployed running instances and those running instances sometimes have complicated topology with multiple running virtual machines (VMs). For example, Figure 1 depicts an example topology pattern for a cloud application where there are 7 VMs, one deployment manager (named Dmgr), four customer nodes, and two database servers (primary/standby DB). The Cloud application deplorer usually needs to apply the emerging fixes (named as iFixes below) and fixpack (a collection of fixes) to the running instances to resolve the exposed problems. It is time consuming task to apply iFixes/FixPack to multiple running VMs especially there are the ordering constraints to apply fixes to multiple VMs. For example in Figure 1, Dmgr must be stopped to apply fix for Dmgr. Dmgr must be restarted first, then a similar process must occur for customer nodes and database servers, one by one.

For an example embodiment of the disclosed method depicting an implementation for the deployment in Figure 1 with seven running VMs, the implementation includes the following:

Protected appliance A : A topology model and VMs provision ordering generation


1.

appliance in Cloud server based on the application pattern created at build time. Protected appliance B: A maintenance orchestration generation appliance in the


2.

master r...