Packet Transfer Search Table Timer-Based Entry Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current packet transfer processing devices with 32-bit architecture can only examine 32 entry areas of a search table at once, leading to inefficient processing and long times when searching for empty areas, and struggle to detect the end of a communication period, resulting in ineffective management of search rules and limited memory utilization.
Innovation Solution
A packet transfer processing device that includes a search table with a validity management table, timers, and a control unit that detects empty entry areas based on timer timeouts, allowing for dynamic management of search conditions and improved memory efficiency by identifying timed-out entries without sequential validity checks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a 32-bit architecture CPU is used to search for empty entry areas in the search table, then the device complexity is reduced, but the processing speed decreases and the time required to search for empty areas increases
Solution Approach 1:
The search table is divided into multiple banks, with each bank containing multiple entry areas. The 2048 entry areas are segmented across 64 banks, allowing parallel search operations. Each bank can be searched independently, enabling the system to process multiple entry areas simultaneously rather than sequentially examining 32 entries at a time.
Solution Approach 2:
The patent introduces a bank dimension to the search table structure, transforming it from a flat single-dimension array to a two-dimensional structure (banks × entry areas per bank). This allows the CPU to access and search multiple banks in parallel, effectively increasing the search capacity from 32 entries to 2048 entries while maintaining 32-bit architecture constraints.
2Device complexity
If the CPU sequentially examines entry areas to detect empty areas in the search table, then the device complexity is reduced, but the time required to detect communication period ends increases
Solution Approach 1:
A timer is associated with each entry area in the search table and starts counting when a packet search condition is set. The timer automatically detects timeout conditions without requiring CPU intervention, preliminarily preparing the information about which entries are still active. This allows the CPU to quickly identify empty areas by checking timer status rather than sequentially examining each entry.
Solution Approach 2:
The timer mechanism serves itself by automatically counting down and generating timeout signals when communication periods end. Each timer independently monitors its associated entry area, eliminating the need for the CPU to actively query or sequentially check each entry's status. The system uses the timers' self-generated timeout information to rapidly identify empty areas for reuse.
3Adaptability or versatility
If a large number of filtering rules are stored in the search table, then the packet filtering capability is improved, but the storage capacity requirement increases
Solution Approach 1:
The search table entries are dynamically managed through a valid/invalid flag mechanism. When a timer times out, the corresponding entry is marked invalid and becomes available for reuse. This dynamic allocation allows the system to support a large number of filtering rules over time, rather than being limited by the fixed number of physical entry areas. The same 2048 entry areas can serve different filtering rules at different times.
Solution Approach 2:
When communication periods end, the corresponding search rules are discarded by marking their entry areas as invalid. These discarded entries are then recovered and made available for storing new filtering rules. This recycling mechanism allows the system to effectively handle a large number of filtering rules sequentially, maximizing the utilization of limited storage capacity while maintaining comprehensive packet filtering capability.
Data Source
AI summary
The end of a communication period can be detected reliably and efficiently by a control unit and validity/invalidity of search rules in a search table can be effectively managed. A timer is provided so as to correspond to each entry area of the search table. Then, under control of a timer managing unit, the timer starts a counting operation whenever information is written in a validity management table or a packet in which a communication parameter matches a packet search condition corresponding to information on the validity management table that is set as valid is received. In addition, when a counting time of the timer exceeds a preset time-out time, information indicating the time-out is stored in a time-out management table.