Multi-Cloud Deployment Controller Cost Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Customers face difficulties in cost-effectively deploying analytics applications across multiple cloud systems due to varying costs of data storage, computation resources, and inter-node communication, making it challenging to optimize deployment configurations in multi-cloud environments.

Innovation Solution

A method and system that utilize a deployment controller and deployment cost analyzer to estimate and minimize the total cost of deploying analytics applications by selecting the most cost-effective computing nodes for storage, computation, and inter-node communication across multiple clouds, considering storage costs, computation resource costs, and data transfer costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If data is stored in storage nodes distributed in different storage clouds and computation is distributed in different clouds, then application deployment flexibility and resource utilization are improved, but data movement costs and deployment complexity increase significantly

Engineering Contradiction:

Improvedeployment flexibilityVSAvoiddeployment complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system changes the parameters of deployment configuration by evaluating multiple cloud environments and selecting optimal storage and computation nodes based on cost, performance, and data locality parameters. This allows flexible deployment across different cloud providers while managing complexity through automated parameter optimization.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces an intermediary deployment optimization system that mediates between the application and multiple cloud providers. This intermediary automatically manages the complexity of distributed deployment by handling data movement coordination, cost calculation, and node selection, thereby improving deployment flexibility without proportionally increasing user-facing complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If data movement services are used for inter-node communications across different clouds, then application execution capability is improved, but data transfer costs become significant

Engineering Contradiction:

Improveapplication execution capabilityVSAvoiddata transfer cost

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The system applies local quality by keeping data and computation locally co-located whenever possible within the same cloud environment. The optimization algorithm prioritizes deployment configurations where storage and computation nodes are in the same cloud region, thereby enabling application execution while minimizing cross-cloud data transfer costs.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent creates equipotential deployment configurations by balancing data location and computation location within the same cloud environment, eliminating potential differences that would drive expensive data movement. This ensures application execution capability is maintained while data transfer costs are minimized by avoiding unnecessary cross-cloud communications.

Inventive Principle:

Principle #12Equipotentiality

3Quantity of substance

If multiple cloud systems are used for deployment, then resource availability and scalability are improved, but cost optimization becomes difficult and confusing

Engineering Contradiction:

Improveresource availabilityVSAvoidcost optimization ease

Core Design Contradiction:

Quantity of substanceVSEase of operation

Solution Approach 1:

The patent implements a universal cost optimization system that works across multiple cloud providers and deployment scenarios. This single system provides multi-functional capabilities including cost calculation, performance evaluation, and automated configuration selection, thereby improving resource availability across clouds while simplifying cost optimization for users.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The deployment optimization system operates autonomously to self-determine the most cost-effective configuration across multiple cloud providers. It automatically evaluates resource availability, calculates costs, and selects optimal deployment configurations without requiring users to manually navigate complex pricing structures, thereby maintaining high resource availability while making cost optimization easy.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS10917463B2Minimizing overhead of applications deployed in multi-clouds

Publication Date: 2021.02.09 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

A computer readable storage medium and methods for distributing an application among computing nodes in a distributed processing system. A method estimates a cost of storing information pertaining to the application on different computing nodes; estimates a cost for computing resources required to execute the application on different computing nodes; estimates a cost of inter-node communication required to execute the application on different computing nodes; and selects at least one computing node to execute the application based on minimizing a total of at least one of the cost estimates.