AI Workload Migration Decisions for Cloud Cost and Latency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Organizations face challenges in optimizing the use of private and public cloud resources to balance performance and cost, with high costs associated with switching providers and transferring large volumes of data, and the need for efficient workload migration strategies.

Innovation Solution

A method and system for monitoring and analyzing metrics using a first computing system to determine migration criteria, initiating workload migration to a third computing system based on cost and latency, and storing pre-migration state information to facilitate seamless transition, utilizing artificial intelligence algorithms for decision-making.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If workload is migrated between cloud providers to optimize cost, then cost is reduced, but data transfer time and switching costs increase

Engineering Contradiction:

ImprovecostVSAvoiddata transfer time

Core Design Contradiction:

Loss of energyVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of workload metrics, cost data, and performance parameters before migration decisions are made. By pre-evaluating multiple cloud providers and preparing migration candidates, the system reduces actual migration time and enables cost optimization without excessive data transfer delays.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors workload performance, cost metrics, and migration outcomes, using this feedback to refine future migration decisions. This closed-loop approach ensures that cost reductions from provider switching do not result in prolonged data transfer times, as the system learns from past migration experiences.

Inventive Principle:

Principle #23Feedback

2Productivity

If private cloud resources are used to maximize performance, then performance is improved, but cost increases

Engineering Contradiction:

ImproveperformanceVSAvoidcost

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The system dynamically adjusts workload placement between private and public cloud resources based on real-time performance requirements and cost conditions. Rather than statically allocating all workloads to private cloud for maximum performance, the system flexibly migrates workloads based on current needs, achieving performance optimization while controlling costs.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system applies different resource allocation strategies to different workloads based on their specific performance requirements and cost sensitivity. Critical performance-intensive workloads are placed on private cloud resources, while less demanding workloads use public cloud resources, optimizing the overall balance between performance and cost.

Inventive Principle:

Principle #3Local quality

3Adaptability or versatility

If public cloud resources are used to handle usage spikes, then scalability is improved, but cost increases

Engineering Contradiction:

ImprovescalabilityVSAvoidcost

Core Design Contradiction:

Adaptability or versatilityVSLoss of energy

Solution Approach 1:

The system uses public cloud resources partially, only for handling usage spikes and overflow workloads rather than for all computing needs. This partial utilization of public cloud resources provides scalability during peak demands while minimizing cost increases by keeping baseline workloads on more cost-effective private resources.

Inventive Principle:

Principle #16Partial or excessive action

4Loss of energy

If workload migration is performed frequently to optimize costs, then cost is reduced, but system stability and performance degradation increase

Engineering Contradiction:

ImprovecostVSAvoidsystem stability

Core Design Contradiction:

Loss of energyVSStability of the object's composition

Solution Approach 1:

The system performs workload migration periodically based on scheduled evaluations and threshold-based triggers rather than continuously or too frequently. This periodic approach allows the system to optimize costs through strategic migrations while maintaining stability by avoiding excessive migration activity that would cause performance degradation and system instability.

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS12517756B2Learning model to optimize automatic workload migration

Publication Date: 2026.01.06 DELL PROD LP

AI summary

Metrics, including cost and latency, corresponding to a public computing system are monitored. A determination to migrate a workload running at donor computing system may be made based on the monitored metrics. A learning model, which may be initialized based on factors from a migration rules engine function, may make recommendations to migrate a workload based on a training data corpus. The learning model may be trained based on manual selections, or acceptances, of recommendations Automatic migration of a workload, under control of the learning model, may be made if recommendations are selected or accepted more frequently than an automatic migration criterion. Different learning models may be used to determine to migrate different corresponding workloads.