Cross-Transmit Table Entry Reading for Network Device Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional routing table lookup systems face reliability issues due to the one-to-one correspondence between interface modules, DDR controllers, and DDR SDRAMs, leading to low reliability and suboptimal bandwidth utilization during multi-interface access.
Innovation Solution
A table entry reading method and apparatus that allows interface modules, controllers, and memories to cross-transmit read commands and table entries, enabling the use of alternative controllers if one fails and allowing for shared bandwidth by not maintaining a one-to-one correspondence, thereby improving reliability and bandwidth utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If interface modules, DDR controllers, and DDR SDRAMs are in a one-to-one correspondence, then the system structure is simple and easy to control, but the reliability is low and bandwidth utilization is suboptimal
Solution Approach 1:
The patent implements a many-to-many mapping relationship where interface modules can access multiple DDR controllers and DDR SDRAMs through a crossing module. This allows any interface module to utilize any available DDR controller and memory, transforming the rigid one-to-one correspondence into a flexible universal access structure that improves reliability while maintaining manageable complexity through standardized crossing module interfaces.
Solution Approach 2:
The crossing module serves as an intermediary between interface modules and DDR controllers, managing the many-to-many mapping relationships. It receives read commands from interface modules, determines appropriate target controllers and memories based on polling rules and correspondence relationships, and routes commands accordingly. This intermediary structure enables flexible resource allocation without requiring direct complex interconnections between all components.
2Productivity
If interface modules, DDR controllers, and DDR SDRAMs are in a one-to-one correspondence, then the control logic is simple, but the bandwidth utilization is suboptimal
Solution Approach 1:
The patent enables multiple interface modules to share access to multiple DDR controllers and memories through the crossing module's many-to-many mapping capability. This allows bandwidth to be dynamically allocated and shared across multiple interfaces, improving overall utilization rates while the crossing module manages the complex mapping relationships through standardized polling rules and correspondence tables.
Solution Approach 2:
The system implements dynamic resource allocation where the crossing module can change the mapping relationships between interface modules and DDR controllers based on real-time conditions. The polling rules and correspondence relationships allow the system to adaptively route read commands to different controllers and memories, enabling dynamic bandwidth sharing and load balancing rather than static one-to-one assignments.
3Reliability
If a DDR controller fails in a one-to-one correspondence system, then the failure is isolated to a single interface module, but the overall system reliability is low
Solution Approach 1:
The patent implements redundancy through many-to-many mapping where each interface module can access multiple DDR controllers and memories. If one controller fails, the crossing module can redirect read commands from affected interface modules to alternative healthy controllers based on the maintained correspondence relationships, providing fault tolerance without requiring complex active failover mechanisms.
Solution Approach 2:
The system prepares for potential failures by maintaining multiple correspondence relationships between interface modules and DDR controllers in advance. The crossing module stores polling rules and multiple mapping paths so that when a controller fails, the system can immediately switch to pre-planned alternative routes without complex real-time decision-making, providing fault tolerance through预先 prepared redundancy.
Data Source
AI summary
This application provides a table entry reading method and apparatus, a network device, and a computer-readable storage medium. The method includes: when a read command is received from an interface module, determining a target controller from a plurality of controllers by polling according to a preset polling rule, and determining a correspondence between the interface module and the target controller; sending the read command to the target controller, so that the target controller obtains a table entry corresponding to the read command from a target memory corresponding to the target controller in a plurality of memories; and receiving the table entry returned by the target controller, and sending the table entry to the interface module based on the correspondence between the interface module and the target controller.