Flexible Meta Model for Cross-Service Observability Data Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Monitoring web services in a distributed environment is challenging due to complex data dependencies and lack of relational awareness, leading to difficulties in associating logged data across different services and risking software application breakdowns.
Innovation Solution
An extensibility platform utilizing a Flexible Meta Model (FMM) for centralized monitoring, enabling interweaving of multiple models in a matrix form to model Metrics, Events, and Traces (MELT) data, with tenant-specific solution packages, and processing observability data based on globally unique identifiers and attribute fields.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If web services are monitored in a discrete and non-centralized fashion, then each web service can be monitored independently, but it becomes difficult to associate logged data across different web services and risks breaking the software application
Solution Approach 1:
The patent merges discrete monitoring approaches into a centralized monitoring system that collects logs from multiple web services into a unified repository. This allows independent monitoring of each service while maintaining the ability to associate logs across services through common identifiers and relationships defined in the extensible data model.
Solution Approach 2:
The patent creates a universal monitoring framework with an extensible data model that can handle multiple types of web services and log formats. The system provides multi-functional capabilities including centralized log collection, correlation across services, and flexible querying, resolving the contradiction between independent monitoring and data association.
2Reliability
If monitoring code is added for one web service, then monitoring capability is improved, but it risks breaking the software application by not accounting for dependencies and overall execution
Solution Approach 1:
The patent segments the monitoring system into independent, loosely-coupled components including log collectors, a centralized repository, and analysis tools. This segmentation allows monitoring code to be added to individual web services without affecting other services, while the centralized repository maintains awareness of dependencies and overall execution through structured data relationships.
Solution Approach 2:
The patent introduces a centralized log repository and extensible data model as intermediaries between web services and monitoring analysis. This intermediary layer absorbs integration complexity, allowing monitoring code to be added to individual services without directly impacting other services or the overall application execution.
3Adaptability or versatility
If an extensible data model is implemented, then customization and extension of monitoring solutions is enabled, but device complexity increases
Solution Approach 1:
The patent implements a dynamic extensible data model where the schema can be customized and extended based on specific monitoring needs. The model allows adding new log types, attributes, and relationships without restructuring the entire system, enabling customization while managing complexity through incremental evolution rather than rigid upfront design.
Data Source
AI summary
According to embodiments of the disclosure, an example method herein may comprise: providing an extensibility platform for observing entities to produce associated observability data; installing flexible meta models within the extensibility platform, wherein the plurality of flexible meta models define the entities, a globally unique identifier of each of the plurality of entities, a type of each of the entities, relationships between the entities, kinds of observability data, and dependencies among the flexible meta models; and processing the observability data obtained within the extensibility platform based on tenant-specific solution packages and the flexible meta models, wherein the observability data is associated to the entities based on external references to corresponding globally unique identifiers of the entities, and wherein the observability data is sourced by sources configured to populate, based on a corresponding observed entity, attribute fields and tenant-specified tag fields according to that corresponding observed entity.