Clustered Gateway Shunting With Hash-Based Packet Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The implementation of shunting clustered gateways in large-scale clusters is complex and resource-intensive, requiring significant memory and table entry capacity, and is typically achieved through cooperative work of multiple servers.
Innovation Solution
A method and apparatus for shunting clustered gateways using a pre-configured port dictionary to generate a hash value for packet routing, selecting a target gateway machine, and forwarding packets based on this hash value, without the need for service-related network address translation rules, utilizing static hash dictionaries and high-speed hardware.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If shunting of clustered gateways is implemented using public network IPs with NAT rules configured on the shunt, then users can select the correct public network IP among multiple public network IPs, but the implementation becomes complex and requires very high memory capacity and table entry capacity, especially in large-scale clusters
Solution Approach 1:
The patent segments the shunting function by introducing dedicated gateway machines that handle specific shunting tasks, separating the complexity from the main shunt device. The shunt forwards packets to gateway machines based on simple criteria, while gateway machines maintain the complex mapping tables and perform detailed shunting logic, thus distributing complexity across multiple specialized components.
Solution Approach 2:
The patent introduces gateway machines as intermediary components between the shunt and the public network. These gateway machines act as mediators that handle the complex NAT rule matching and public IP selection, allowing the main shunt to operate with simpler logic while still achieving correct routing through the intermediary gateway layer.
2Productivity
If shunting of large-scale clustered gateways is achieved through cooperative work of a large number of servers, then the system can handle large-scale clusters, but the memory capacity and table entry capacity requirements become extremely high
Solution Approach 1:
The patent divides the large-scale cluster into multiple gateway machines, each responsible for specific subsets of NAT rules and public IP mappings. This segmentation allows the system to handle large-scale clusters by distributing the memory and table entry requirements across multiple machines rather than concentrating them in a single shunt, making the system scalable without linearly increasing individual device resources.
Solution Approach 2:
The patent introduces a new dimensional layer in the network architecture by adding gateway machines as an intermediate layer between the shunt and end devices. This dimensional change allows the system to scale horizontally by adding more gateway machines rather than vertically increasing the resources of existing devices, thus handling large-scale clusters with distributed rather than concentrated resources.
3Reliability
If all NAT rules in a cluster are configured on the shunt to ensure correct public network IP selection, then accurate routing is achieved, but the shunt becomes a bottleneck with high resource consumption
Solution Approach 1:
The patent extracts the complex NAT rule matching and public IP selection functions from the shunt and relocates them to dedicated gateway machines. The shunt retains only the essential packet forwarding function, while the extracted shunting logic resides on gateway machines that have specialized hardware and software optimized for these tasks, thereby improving overall processing efficiency and removing the bottleneck from the shunt.
Solution Approach 2:
The patent enables gateway machines to independently perform shunting decisions for their assigned subsets of rules without requiring constant consultation with the main shunt. Each gateway machine maintains its own NAT rule tables and can autonomously determine the correct public IP for packets it receives, allowing parallel processing and eliminating the shunt as a single point of decision-making bottleneck.
Data Source
AI summary
The present disclosure provides methods for shunting clustered gateways, which relate to the field of computer technologies, and in particular, relate to the technical field of data transmission. A specific implementation solution is: in response to receiving a first packet sent by a target internal network machine, a first hash value is acquired, wherein the first hash value is generated by a shunt of a public network gateway cluster, and the shunt is configured to perform a hash calculation on the first packet based on a pre-configured port dictionary; a target gateway machine is selected from the public network gateway cluster according to the first hash value; and the first packet is sent to the target gateway machine.