Adaptive Application Placement Bot for Cloud Environments

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current application placement decision support systems in cloud environments rely on static, non-customizable rules that do not directly correlate with application attributes, leading to ineffective placement decisions as environments evolve.

Innovation Solution

An adaptive application placement management bot that obtains application metadata, correlates attributes with a dynamic master attributes table using machine learning to identify new attributes and generate customized rules for determining optimal placement across various environments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If static rules with fixed parameters are used for application placement decisions, then the system structure is simple and easy to implement, but the placement effectiveness decreases over time as application environments evolve

Engineering Contradiction:

Improveease of implementationVSAvoidplacement effectiveness

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The patent transforms the static rule system into a dynamic one by implementing machine learning models that continuously learn from application performance data and environment changes. The system adapts placement rules over time based on observed patterns, ensuring placement effectiveness is maintained as environments evolve while retaining implementation feasibility through automated learning processes.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent incorporates feedback mechanisms where placement decisions are monitored, evaluated, and fed back into the system to refine future decisions. Performance metrics from deployed applications are collected and used to adjust placement rules, creating a closed-loop system that improves reliability over time while maintaining operational simplicity through automated feedback processing.

Inventive Principle:

Principle #23Feedback

2Ease of operation

If predefined rules with fixed parameters are applied, then the decision support system is easy to operate, but the rules are not customizable or adaptable to different application requirements

Engineering Contradiction:

Improveease of operationVSAvoidcustomizability

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The patent enables the system to automatically adapt to different application requirements through self-learning machine learning models. The system autonomously analyzes application-specific characteristics and adjusts placement rules without requiring manual customization, maintaining ease of operation while achieving high adaptability through automated self-adjustment mechanisms.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent implements dynamic parameter adjustment where placement rules are not fixed but can be modified based on learned patterns from application performance data. The system changes relevant parameters automatically to suit different application types and requirements, providing both ease of operation and high customizability through adaptive parameter transformation.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If static parameters are used in placement rules, then the rule set is simple to maintain, but the parameters do not directly correlate with imported application attributes

Engineering Contradiction:

Improverule complexityVSAvoidattribute correlation accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces the mechanical system of manually defined static parameters with an intelligent system using machine learning models. These models automatically identify and establish correlations between application attributes and placement parameters, achieving high measurement precision without increasing operational complexity since the correlation process is automated through learning algorithms.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces machine learning models as intermediaries between imported application attributes and placement rules. These intermediary models automatically process and correlate attributes with appropriate placement parameters, maintaining simple rule structures while achieving accurate attribute correlation through the intermediary learning layer that translates between different data representations.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS10445080B2Methods for adaptive placement of applications and devices thereof

Publication Date: 2019.10.15 WIPRO LTD

AI summary

Methods, application placement management bot, and non-transitory computer readable media that obtain application meta data for a plurality of applications. The application meta data comprises application requirements and associated application attributes for each of the applications. A first subset of the application attributes is correlated with master attributes in a master attributes table based on a first set of keywords matching a stored second set of keywords mapped to the master attributes. The first set of keywords corresponds to one or more of the application requirements associated with the first subset of the application attributes. Rule set(s) are obtained that comprise customized rule(s) based on the master attributes. The rule set(s) are applied to the application meta data based on the correlation to determine a placement of each of the applications in at least one of a plurality of environments. An indication of the determined placement is output.