ML Microservice Platform for Cloud Infrastructure Placement

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current cloud computing environments lack efficient and automated mechanisms for provisioning applications across both Virtual Machine (VM) and container-based infrastructures, often resulting in suboptimal resource utilization and decision-making between hypervisor and container approaches.

Innovation Solution

An automated provisioning system coupled with a machine learning-based microservice setup platform that communicates with VM and container provisioners to determine cluster data, execute policy rules, and generate recommendations for optimal infrastructure assignment based on customer demands, enabling intelligent geographical placement and smarter archival and decommissioning processes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated provisioning is implemented across both VM and container infrastructures, then resource utilization and provisioning efficiency are improved, but system complexity increases

Engineering Contradiction:

Improveprovisioning efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces a machine learning-based intermediary system that acts as a mediator between provisioning requests and the underlying VM/container infrastructures. This ML system analyzes workload characteristics, predicts optimal placement decisions, and manages the complexity of coordinating multiple provisioners, thereby improving provisioning efficiency without directly increasing operational complexity for end users.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces manual or rule-based mechanical provisioning decisions with a machine learning system that automatically analyzes patterns and makes intelligent decisions. The ML model substitutes complex manual configuration processes with automated predictive algorithms, improving productivity while the system learns and adapts to reduce operational complexity over time.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If machine learning-based recommendations are used for infrastructure assignment, then decision accuracy and resource optimization are improved, but processing time and computational resources increase

Engineering Contradiction:

Improvedecision accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements preliminary action by pre-training machine learning models on historical provisioning data and workload patterns before actual provisioning decisions are needed. The system performs offline training and validation to establish predictive models, so that during actual provisioning, the pre-trained models can quickly make accurate recommendations without requiring extensive real-time computation, thus reducing processing time while maintaining high decision accuracy.

Inventive Principle:

Principle #10Preliminary action

3Adaptability or versatility

If intelligent geographical placement and archival processes are implemented, then cost optimization and resource utilization are improved, but system complexity and implementation difficulty increase

Engineering Contradiction:

Improveresource optimization capabilityVSAvoidimplementation difficulty

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The patent applies universality by designing a multi-functional machine learning platform that handles multiple provisioning scenarios (VM and container placement, geographical distribution, archival decisions) through a single unified system. This universal platform can adapt to different infrastructure types and provisioning needs without requiring separate specialized systems, thereby improving resource optimization capability while actually reducing implementation difficulty through consolidation rather than proliferation of separate components.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11630685B2Hypervisor and container placement and cost optimization utilizing machine learning

Publication Date: 2023.04.18 SAP SE

AI summary

According to some embodiments, an automated provisioning system may receive a customer demand associated with an application to be executed in a cloud-based computing environment. The automated provisioning system may include a process allocator to communicate with Virtual Machine (“VM”) and container provisioners and determine cluster data. A machine learning based microservice setup platform, coupled to the automated provisioning system, may receive the cluster data and information about the customer demand. The machine learning based microservice setup platform may then execute policy rules based on the cluster data (and information about the customer demand) and generate a recommendation for the customer demand. The automated provisioning system may then assign the customer demand to one of a VM-based infrastructure and a container-based infrastructure in accordance with the recommendation generated by the machine learning based microservice setup platform.