Kafka-Based Cellular Observability for Centralized Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Telecommunications companies face challenges in cost-effectively expanding their network infrastructure while improving user experience due to the need for extensive observability in radio access networks, which often require specialized hardware and software.
Innovation Solution
A centralized management system utilizing a containerized application framework, such as kubernetes clusters, to collect and manage data across all domains of a cellular network, incorporating an observability layer that stores data at a common source and processes it in real-time for distribution to applications, with a processor managing domains directly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If specialized hardware and software are used in radio access networks to ensure proper system operation, then system reliability is improved, but device complexity and cost increase
Solution Approach 1:
The patent segments the radio access network into multiple network domains (EPC domain, 5G core domain, RAN domain) with dedicated observability collectors for each domain. This segmentation allows specialized monitoring for each domain while managing overall system complexity through modular architecture.
Solution Approach 2:
The patent introduces an intermediary data bus (Kafka bus) that mediates between multiple observability collectors and a centralized configuration management system. This intermediary layer simplifies data collection across domains by providing a standardized interface, reducing the complexity of direct point-to-point connections.
2Loss of information
If data is collected and stored at multiple distributed locations across network domains, then data availability is improved, but data management complexity increases
Solution Approach 1:
The patent merges multiple distributed data sources into a unified configuration management system that centralizes configuration data while maintaining domain-specific collectors. The data bus combines data from EPC, 5G core, and RAN domains into a single streaming platform, reducing management complexity through consolidation.
Solution Approach 2:
The patent creates a universal data bus (Kafka bus) that serves multiple functions: collecting data from different domains, storing configuration information, enabling real-time streaming, and supporting various applications. This multi-functional approach reduces the need for separate systems for each function.
3Productivity
If real-time data processing is implemented across all network domains, then user experience is improved, but processing requirements and energy consumption increase
Solution Approach 1:
The patent implements preliminary action by pre-configuring observability collectors in each network domain to automatically collect and publish data to the data bus. Configuration information is pre-stored in the centralized system, enabling real-time processing without ad-hoc data gathering, thus reducing processing overhead.
Solution Approach 2:
The patent establishes continuous data collection and streaming across all network domains through the Kafka bus, ensuring uninterrupted real-time monitoring. This continuous action eliminates periodic sampling overhead and enables efficient event-driven processing only when changes occur.
Data Source
AI summary
A system includes domains comprising systems generating data from a cellular network employing clusters created using a containerized application, and an observability (OBF) layer configured to collect the data over all of the domains and store the data at a common source. The system also includes a streaming platform data store configured for ingesting and processing streaming data in real-time and distributing the data to applications requesting such data. The streaming platform extends completely over the OBF layer so that all data collected by the OBF layer is available to the streaming platform data store. The system also includes a processor configured to collect data at the common source and manage the domains directly using the data collected at the common source.