AI Microservice Bundling for Shared Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing microservice architectures struggle to efficiently manage resource sharing among microservices, leading to inefficiencies and potential downtime due to independent scaling and security challenges.
Innovation Solution
A processor utilizes an AI model to analyze features of multiple microservices, classifying subsets as bundles that can share resources based on utilization contexts, traffic patterns, resource needs, and criticality, thereby optimizing resource allocation and reducing downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If microservices are deployed independently to maintain security boundaries and enable independent scaling, then security and flexibility are improved, but resource utilization efficiency deteriorates
Solution Approach 1:
The patent groups microservices into bundles based on their utilization contexts and resource needs, allowing multiple microservices to share common resources such as databases, message queues, and infrastructure. This merging approach maintains the independent deployment and scaling capabilities of individual microservices while improving overall resource utilization efficiency by eliminating redundancy across service boundaries.
2Loss of energy
If microservices are bundled to share resources and improve efficiency, then resource utilization is improved, but system complexity increases
Solution Approach 1:
The patent segments the microservice architecture into distinct bundles, where each bundle is a logical grouping of microservices that share common resources and utilization contexts. This segmentation allows the system to manage complexity by organizing microservices into manageable units rather than treating them as completely independent entities, thereby improving resource efficiency without proportionally increasing overall system complexity.
Solution Approach 2:
The patent introduces an AI-based classification system as an intermediary that automatically determines which microservices should be bundled together based on their utilization patterns and resource needs. This intermediary layer simplifies the bundling process and reduces manual configuration complexity by using machine learning models to make intelligent grouping decisions.
3Adaptability or versatility
If independent scaling is implemented for each microservice, then flexibility and security are improved, but downtime increases due to lack of coordinated resource allocation
Solution Approach 1:
The patent implements dynamic resource allocation at the bundle level, where resources are allocated and scaled based on the collective needs of microservices within a bundle rather than independently for each service. This dynamic approach allows the system to maintain flexibility in scaling individual microservices while coordinating resource allocation across bundles to prevent downtime and improve overall system reliability.
Data Source
AI summary
A processor may receive data regarding two or more microservices. The processor may identify, using an AI model, features of the two or more microservices. The processor may analyze the features of the two or more microservices relating to utilization contexts of the two or more microservices. The processor may classify a subset of the two or more microservices as a microservice bundle based on the analyzed features, where the microservice bundle includes two or more microservices configured to share microservice resources. The processor may output the classification of the microservice bundle.