Browse Prior Art Database

System and method to dynamically derive and apply properties to an appliance and use the same to dynamically control provisioning of workloads

IP.com Disclosure Number: IPCOM000236967D
Publication Date: 2014-May-23
Document File: 7 page(s) / 191K

Publishing Venue

The IP.com Prior Art Database

Abstract

In a Cloud Environment, there is a proliferation of appliances which makes it difficult for users to find an appropriate appliance for a specific purpose. Also, there are appliances which are prone to fail at various life-cycle stages such as at the time of provisioning, deploying, booting, authentication, capture etc. However, there is no automated and intelligent way to classify appliances and also to identify troublesome appliances. This classification and filtering of appliances is useful to indicate the users as well as to use this information by provisioning apparatus. This invention publication presents a novel mechanism to solve this problem in a Cloud Environment.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 31% of the total text.

Page 01 of 7

System and method to dynamically derive and apply properties to an appliance and use the same to dynamically control provisioning of workloads

In a typical private cloud environment, there can be numerous appliances. Over the period when these appliances become more, cloud users always find it difficult to select best appliance for a specific purpose for deployment. For example: If user is looking for a VM for some critical work, he/she wants a VM which is most reliable or if a user wants a VM very quickly for some specific use, for very short duration he wants an appliance which deploys very fast and ready for use but the problem is how to find such an appliance in a cloud environment?

In a cloud environment, appliances can have many characteristics which reveal the use for the corresponding VM like, some appliance when deployed,


 VM's are used for months and they work seamlessly throughout


 VM's are used for shorter duration of time and then destroyed


 Deployment mostly fails


 Deployment never failed


 VM always captures successfully


 VM never boots


 VM starts but login fails


 Deploy takes lot of time


 Deploys are successful only when used with certain flavors (Tiny/Medium/large)

But the problem is how user would know which appliance is good or bad. In a nutshell, it is really critical to select a right appliance for deployment to save energy, cost and time. Currently, there are some ways to indicate these properties for Appliances, however, they have limitations such as:


1. Mostly very primitive i.e. they only depict appliances with some very high level attributes.

2. These attributes are updated by Admin / users manually
3. There is no decent way to identify and update these attributes dynamically and reliably in an automated manner.

4. These attribute are left for the user to make choice, if at all they are available. No way to affect / consume these details in actual scheduling process automatically.

5. No feedback loop to make the whole cloud working better over deployments, dynamically and automated manner.

This disclosure aims to provide a way to derive and apply properties or characteristics to an appliance with various deployment and corresponding VM usage characteristics based on historical analytical data in a cloud environment in a dynamic and automated manner which is accurate, reliable and relevant to users.

According to the disclosure, it is proposed to dynamically derive and apply properties to an appliance based on the historic analytical data collected from the past usage of the appliance.

Data Point sources:


For each of appliance when deployed, following would be the source of data collection:

1. Time, when appliance deployed 2. Time, when appliance destroyed 3. Reasons, if user destroys a VM: a. Collect user inputs while he/she destroys a VM:
i. Work flow is successfully finished
ii. Couldn't boot VM
iii. Couldn't login to the VM
iv. IO operations didn't worked

1


Page 02 of 7


v. Any othe...