DCN Packet Forwarding in OTN Payloads to Cut CPU Delay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing optical transport networks (OTN) face challenges with insufficient bandwidth for control information and inefficient CPU processing due to the centralized management of network elements, leading to high CPU consumption and processing delays.
Innovation Solution
Implementing a hardware processing unit to identify and forward direct DCN packets directly to their target devices within the payload area of data frames, reducing the load on CPUs and expanding the transmission bandwidth for DCN packets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If DCN packets are sent to CPU for processing, then packets can be identified and processed, but data packet processing delay increases and CPU reading resources are greatly consumed
Solution Approach 1:
The patent segments packet processing into two paths: direct packets are handled by hardware processing units for fast forwarding, while only indirect packets requiring actual processing are sent to the CPU. This segmentation reduces CPU involvement and processing delay for the majority of packets that simply need to be forwarded.
Solution Approach 2:
The patent introduces a hardware processing unit as an intermediary between packet reception and CPU processing. This intermediary quickly identifies direct packets and forwards them without CPU intervention, reserving CPU resources for packets that truly require processing while maintaining accurate packet handling.
2Reliability
If control information is transferred through GCC overhead field, then management and monitoring functions are provided, but bandwidth occupied is limited by rate level of ODUk
Solution Approach 1:
The patent moves DCN packet transmission from the traditional GCC overhead field to the payload area of data frames, effectively changing the dimensional space for control information transmission. This allows utilization of much larger bandwidth available in the payload area while maintaining all management and monitoring functions.
Solution Approach 2:
The patent makes the payload area serve dual purposes: carrying service data and carrying DCN control packets. This multi-functionality increases the bandwidth available for control information without requiring separate dedicated overhead fields, thereby resolving the bandwidth limitation.
3Measurement precision
If all DCN packets are processed by CPU, then packet processing is accurate, but CPU processing performance degrades
Solution Approach 1:
The patent segments packet processing responsibilities between hardware processing units and CPU. Hardware units handle direct packets through fast forwarding with high accuracy, while CPU processes only indirect packets. This segmentation maintains processing accuracy while significantly improving overall system productivity by reducing CPU bottleneck.
Solution Approach 2:
The patent replaces CPU-based software processing with hardware-based processing for direct packets. This substitution uses hardware's inherent speed and efficiency to handle packet forwarding, maintaining accuracy while dramatically improving processing throughput and reducing CPU load.
Data Source
AI summary
This application provides example packet processing methods, network devices, and systems. One example method includes obtaining, by a first network device, a first data communication network (DCN) packet by using a hardware processor, where the first DCN packet is carried in a payload area of a first data frame, and the first DCN packet includes a target device identifier. The first network device determines, based on the target device identifier, by using the hardware processor, whether the first DCN packet is a direct packet. In response to determining that the first DCN packet is a direct packet, the first network device sends the first DCN packet to a second network device, where the second network device corresponds to the target device identifier.