Browse Prior Art Database

Framework for sharing and updating accumulated knowledge across cognitive system deployments

IP.com Disclosure Number: IPCOM000249576D
Publication Date: 2017-Mar-03
Document File: 3 page(s) / 54K

Publishing Venue

The IP.com Prior Art Database

Abstract

When a truly dynamic resource optimization and configuration is achieved through cognitive server provisioning, it becomes important to share this knowledge across multiple deployments. Cross sharing of rules and knowledge relevant to server provisioning, and building dynamism around the accumulation of the knowledge and also in the ingestion of this knowledge by the systems and propagation of a derivative from such accumulated knowledge from different systems makes the core of this framework.

Cognitive server provisioning sets the stage for the corpus accumulated dynamically by a single cognitive system by dynamically observing the environment and recording the performance of the servers in correlation with factors relating to configuration and scaling.

A close match to what is proposed here are the HOT - Heat Orchestration Templates from OpenStack that lets people author a template and make it available for public consumption. But this idea here talks about a framework for dynamically sharing such information across several deployments keeping in mind that the updating to this information is also dynamic

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 61% of the total text.

1

Framework for sharing and updating accumulated knowledge across cognitive system deployments

It describes a framework for cognitive systems to share their accumulated on the job learning dynamically with other such cognitive systems. The following are the dynamic contents shared as part of the framework:

1. Learning related to the configuration information of servers 2. Learning related to how much a certain configuration of server has scaled with respect to memory, processor, network, storage etc 3. Learning related to a business specific configuration and its changing needs 4. User fed learning 5. Learning based on priority, approval status, relevance to business

Following are some of the categories pertaining to the content: 1. Information category: Eg: HowTo, static industry knowledge, dynamic knowledge 2. Relevance to corporate: Intra, Inter corporate or global 3. Relevance to personas: Admin, IT operations Head, Sales Executive

Some of the actions relevant to the content sharing: 1. Sync up between multiple cognitive system deployments 2. Schedule of sync up of each category of data 3. Ingestion, approval, curation

In summary, the framework will be able to do all of the actions, on the content and the categories mentioned above.

The advantages of this idea over the existing HOT templates or some other similar methodology would be the following: 1. The generation of new knowledge are dynamic based on the current changing environment of servers and their configuration...