Data Exchange System Hardware Acceleration for PON CPU Offloading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional Passive Optical Network (PON) systems face high CPU occupancy rates and processing speed limitations due to software-based wireless data processing, which negatively impact the performance of Optical Network Units (ONUs) and their ability to support multiple services.
Innovation Solution
Implementing a data exchange system where the CPU configures data destination ports, enabling the acceleration module to handle data exchange between the DDR and ONU, thereby processing Ethernet data through hardware modules and reducing CPU occupancy rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If wireless data is processed through CPU software acceleration, then data processing can be performed, but CPU occupancy rate increases and processing speed decreases
Solution Approach 1:
The patent replaces CPU software-based wireless data processing with a dedicated hardware processing module. This substitution moves the processing function from the general-purpose CPU to specialized hardware circuitry, enabling parallel processing and significantly reducing CPU occupancy while maintaining or improving processing speed.
Solution Approach 2:
The patent segments the data processing function by separating wireless data processing from general CPU operations. A dedicated processing module is created to handle wireless data independently, allowing the CPU to focus on other tasks and reducing overall system bottlenecks.
2Adaptability or versatility
If CPU handles more services, then service quantity increases, but CPU occupancy rate increases and adversely affects other services
Solution Approach 1:
The patent segments service processing by creating dedicated hardware modules for wireless data processing. This allows the system to support multiple services simultaneously without concentrating all processing load on the CPU, thereby maintaining high service versatility while preserving overall system performance.
Solution Approach 2:
The dedicated processing module is designed to handle wireless data processing universally, freeing the CPU to support multiple other services. This multi-functional architecture enables the system to provide diverse services without the CPU becoming a bottleneck.
3Productivity
If CPU processing capability is increased to handle wireless data, then wireless service performance improves, but device complexity and cost increase
Solution Approach 1:
Instead of increasing CPU processing capability through software optimization or higher-spec CPU hardware, the patent substitutes a dedicated hardware processing module. This approach achieves improved wireless service performance without the complexity and cost associated with upgrading CPU capabilities.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention discloses a data exchange system, which includes a CPU, a DDR, an acceleration module, an ONU, and a wireless network device. The acceleration module reads first Ethernet data and a first descriptor in the DDR to send to the ONU; establishes a MAC address table according to the received first Ethernet data and first descriptor sent by the ONU; modifies the first descriptor by querying the MAC address table, and sends the first Ethernet data and the first descriptor to the ONU. The CPU modifies the first descriptor by configuring a data destination port when the data destination port in the first Ethernet data and the first descriptor, which are received by the ONU, is unknown. The ONU sends the first Ethernet data to the data destination port specified by the first descriptor after processing the first Ethernet data. The present invention also discloses a data exchange method and a computer storage medium.