Service Mesh Across Multiple COIs Without Central Catalog Dependency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies fail to effectively address scalability, interoperability, and security challenges in implementing enterprise-scale service communication systems, with existing service communication systems, and existing technologies require a centralized service catalog, which are not effective in addressing the scalability and interoperability challenges in implementing enterprise-scale communication systems, with existing technologies fail to effectively address scalability, interoperability, and security challenges in implementing enterprise-scale service communication systems, with existing technologies requiring a centralized service catalog, which creates a single point of failure and global runtime dependencies, and security challenges in cross-service authentication and trust management.
Innovation Solution
Implementing a distributed mesh system based on multiple loosely-coupled meshes, each isolated to its own container orchestration instance (COI), with a respective mesh controller for automatic configuration propagation, a DNS controller for service registration and discovery, and an identity framework controller for maintaining a trust domain, using Mutual Transport Layer Security (mTLS) for secure communication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a centralized service catalog is used, then service communication can be managed centrally, but scalability and interoperability challenges arise
Solution Approach 1:
The patent divides the centralized service catalog into distributed service catalogs at each container orchestration instance (COI). Each COI maintains its own service catalog, eliminating the single point of failure and reducing global runtime dependencies. This segmentation enables better scalability and interoperability while preserving centralized management capabilities through the mesh controller that propagates configurations across all COIs.
2Ease of operation
If a centralized service catalog is used, then service communication can be managed centrally, but a single point of failure is created
Solution Approach 1:
The patent segments the centralized service catalog into distributed service catalogs at each COI. This eliminates the single point of failure by ensuring that each COI can operate independently with its own service catalog. The mesh controller ensures consistency across all COIs by propagating configuration changes, but no single COI or centralized catalog can fail the entire system.
3Ease of operation
If a centralized service catalog is used, then service communication can be managed centrally, but global runtime dependencies are created
Solution Approach 1:
The patent segments the service catalog into local COI-specific catalogs, eliminating global runtime dependencies. Each COI manages its own service catalog independently, allowing for localized failures and updates without affecting other COIs. The mesh controller coordinates configuration propagation across COIs, but this does not create global runtime dependencies since each COI operates autonomously with its own catalog.
4Adaptability or versatility
If multiple COIs are used for scaling, then scalability is improved, but configuration propagation and trust management become complex
Solution Approach 1:
The patent segments trust management into isolated trust domains at each COI. Each COI maintains its own trust domain with its own root cryptographic certificate, eliminating the need for complex cross-verification between COIs. The mesh controller propagates configurations across COIs, but trust management remains localized to each COI's trust domain, simplifying the overall system complexity while enabling multi-COI scaling.
5Adaptability or versatility
If control planes are recreated on-demand, then system flexibility is improved, but operational complexity increases
Solution Approach 1:
The patent segments the control plane into isolated COI-specific control planes. Each COI can independently recreate its control plane without affecting other COIs, enabling on-demand recreation for maintenance, upgrades, or failure recovery. The mesh controller coordinates this process across all COIs, but the isolation of control planes to each COI reduces operational complexity compared to managing a single centralized control plane.
Data Source
AI summary
A communication services platform implemented by a distributed computing system may include a plurality of container orchestration instances (COIs), such that a first COI of the plurality of COIs includes a plurality of running instances of a first service accessible via a first mesh proxy and a first mesh controller. The first mesh controller is configured to: publish a first service configuration record for the first service, receive a second service configuration record for a second service running in a second COI of the plurality of COIs, and translate the second service configuration record into one or more custom resource definitions to allow upstream service calls by the second service to the first service.