NRDU Data Path Simulation Server for Multi-Gbps 5G Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for testing data throughput capacity in 5G communication networks are limited by their dependency on Linux IP stack over socket-based interfaces, which restricts achieving multi-Gbps throughput and require additional fast path libraries and drivers, leading to inefficiencies and resource dependencies.
Innovation Solution
A new radio distributed unit (NRDU) data path simulation server is introduced, utilizing a processor with a test controller, DU simulator, traffic generator, IPSEC Manager, scheduler, and UP controller to optimize data throughput capacity by scheduling threads without contention, employing memory allocation and reuse, and simulating IPSEC functionality, all while avoiding dependencies on external fast path libraries and drivers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If Linux IP stack over socket-based interfaces is used, then ease of operation is improved, but data throughput capacity is limited and cannot achieve multi-Gbps
Solution Approach 1:
The system segments the data path processing into multiple independent threads, each handling specific protocol layers (PDCP, RLC, MAC). This allows parallel processing of data packets across multiple threads simultaneously, achieving multi-Gbps throughput while maintaining socket-based interface simplicity for operation.
Solution Approach 2:
The patent introduces a new dimension of thread-level parallelism and memory pool management on top of the traditional socket-based Linux IP stack. By adding this dimensional layer of concurrent thread execution with dedicated memory pools, the system achieves multi-Gbps throughput without sacrificing the operational simplicity of socket interfaces.
2Productivity
If fast path libraries and drivers are used, then data throughput capacity is improved, but device complexity increases due to additional dependencies
Solution Approach 1:
The patent introduces an intermediary layer consisting of thread pools and memory pool managers that sit between the socket-based Linux IP stack and the application layer. This intermediary enables multi-Gbps throughput through efficient memory management and parallel processing without requiring external fast path libraries or proprietary drivers, thus avoiding additional device complexity.
Solution Approach 2:
The system implements self-service through internal thread pools and memory pool management mechanisms that automatically handle data path processing. The threads manage their own memory allocation and data flow without external fast path libraries, making the system self-sufficient and avoiding additional dependencies that would increase device complexity.
3Productivity
If multiple threads are used for data traffic generation, then data throughput capacity is improved, but resource requirements increase
Solution Approach 1:
The system performs preliminary action by pre-allocating memory pools to each thread before data traffic generation begins. This upfront memory allocation eliminates the need for dynamic memory allocation during runtime, reducing overhead and allowing multiple threads to operate efficiently with predictable resource usage, thereby achieving multi-Gbps throughput without proportionally increasing resource requirements.
Solution Approach 2:
The patent implements efficient memory management through discarding and recovering mechanisms where memory blocks are reused across threads. When a thread completes processing, its allocated memory is returned to the pool for reuse by other threads, reducing total memory requirements while maintaining high throughput across multiple concurrent threads.
Data Source
AI summary
Embodiments are directed to a NRDU data path simulation server for testing data throughput capacity in 5G communication network. The NRDU data path simulation server includes a memory comprising a plurality of memory blocks, a processor and a communicator. The processor includes a test controller to receive testing profile for testing the data throughput capacity of NGCU, a DU simulator to generate multiple simulated DUs and multiple simulated UEs based on the testing profile. The processor includes traffic generators to generate multiple threads of data traffic for testing the data throughput capacity of the and a scheduler to schedule each thread to access entities of the simulated DU for optimizing the data throughput capacity and a UP controller to optimize the data throughput capacity of the NRDU data path simulation server by converging a data path of the simulated UEs and a data path of the simulated DUs.