Distributed Delivery Nodes for Service Chain Packet Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In current data center networks, the efficiency of data packet processing is low due to the need for data packets to pass through a delivery node repeatedly, especially when a service chain includes a large number of service nodes, leading to centralized control and reduced processing efficiency.
Innovation Solution
A data packet processing method where a second service node receives an uplink data packet from a first service node, performs service processing, deletes service node information when it determines it is the last node in the uplink direction, and sends the packet to a second delivery node, which then sends it to the destination, eliminating the need for multiple passes through delivery nodes by using distributed delivery nodes for ingress and egress processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If data packets pass through a delivery node repeatedly for centralized control in service chains, then centralized control is achieved, but data packet processing efficiency deteriorates
Solution Approach 1:
The patent divides the centralized delivery node function into distributed delivery nodes. Each delivery node independently performs service chain processing for data packets within its jurisdiction, eliminating the need for repeated centralized control. This segmentation allows parallel processing across multiple delivery nodes, significantly improving data packet processing efficiency while maintaining service chain control capabilities.
2Adaptability or versatility
If data packets pass through multiple service nodes in a service chain, then service processing requirements are met, but the number of passes through delivery nodes increases
Solution Approach 1:
The patent performs preliminary configuration of service chain paths and delivery node routing before data packet transmission. Delivery nodes are pre-configured with service chain information, allowing them to directly forward packets through the appropriate service nodes without repeated control queries. This preliminary setup reduces transmission delays while maintaining the required service processing sequence.
3Adaptability or versatility
If a service chain includes a large quantity of service nodes, then service functionality is enhanced, but data packet processing efficiency deteriorates due to repeated delivery node passes
Solution Approach 1:
The patent introduces a new dimensional approach by organizing delivery nodes in a distributed hierarchical structure rather than a single centralized bottleneck. Data packets can be routed through multiple service nodes in parallel across different delivery node domains, transforming the linear processing path into a multi-dimensional routing structure that scales with service chain complexity without proportionally increasing delivery node pass-throughs.
Data Source
AI summary
The present disclosure discloses a data packet processing method, a service node, and a delivery node, to improve the data packet processing efficiency. A second service node receives a first uplink data packet sent by a first service node. The second service node performs service processing on the first uplink data packet to obtain a second uplink data packet. When the second service node determines, according to service node information, that the second service node is the last service node in an uplink direction of a service chain, the second service node deletes the service node information from the second uplink data packet, to obtain a third uplink data packet, and sends the third uplink data packet to a second delivery node, where the second delivery node is an uplink egress delivery node of the service chain.