Metrics Aggregation for Distributed Software Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Collecting and aggregating metrics for software products deployed as multiple software service instances across distributed computing nodes is challenging, especially when different services handle metrics differently and operate in a highly available environment, making it difficult to determine collective deployment, operation, and usage metrics.
Innovation Solution
Implementing a metrics collector system that allows software service instances to locally calculate metrics and aggregate them based on service identifiers, timestamps, and metric names, using a metric provider library that enables standard reporting and encryption for secure data transmission, facilitating aggregation across different computing nodes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If metrics are collected from multiple distributed software service instances using different collection methods, then comprehensive metrics coverage is improved, but system complexity and difficulty of aggregation increase
Solution Approach 1:
The patent implements a universal metrics collection framework that can handle multiple types of software services (streaming, batch, interactive) through a common interface and aggregation mechanism. The system uses standardized metric names, formats, and collection protocols that work across diverse service types, eliminating the need for service-specific collection logic while maintaining comprehensive metrics coverage.
Solution Approach 2:
The patent introduces an intermediary aggregation layer that sits between distributed software service instances and the metrics collection system. This intermediary standardizes and normalizes metrics from different services before aggregation, handling format conversion, deduplication, and conflict resolution. The intermediary acts as a mediator that translates diverse service metrics into a unified format suitable for centralized aggregation and analysis.
2Reliability
If software services are scaled up across multiple computing nodes for high availability, then service reliability is improved, but metrics aggregation difficulty increases
Solution Approach 1:
The patent segments the metrics collection and aggregation process into distributed components, with each software service instance collecting and pre-processing its own metrics locally. This segmentation allows each node to independently manage its metrics while the overall system maintains a unified view through standardized aggregation protocols. The segmented approach scales linearly with the number of service instances, maintaining aggregation efficiency even as system size grows.
3Adaptability or versatility
If different software products use different metrics collection methods, then product-specific optimization is improved, but unified aggregation capability deteriorates
Solution Approach 1:
The patent allows each software service to maintain its own optimized metrics collection methodology tailored to its specific needs and characteristics. Each service can implement custom collection logic, filtering, and preprocessing locally while adhering to a standardized interface for exporting metrics. This local quality approach enables product-specific optimization without compromising unified aggregation, as the standardized interface ensures compatibility across diverse services.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Various embodiments disclosed herein relate to collection and aggregation of metrics for software services operating across a plurality of computing nodes.