Dynamic TCAM Record Allocation for Routing Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing TCAM systems face performance degradation due to the time-consuming process of keeping entries sorted during routing updates, especially when determining a suitable default block size is difficult, leading to CPU overload and delayed data path forwarding.
Innovation Solution
A dynamic allocation method for TCAM records into clusters based on IP address prefix lengths, where records are initially allocated to specific clusters and a free space cluster, with adjustments made based on thresholds to ensure efficient growth and shrinkage of clusters as needed, allowing for background processing to utilize idle CPU resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If records are allocated to fixed size clusters in TCAM, then insertion of new records becomes simpler, but CPU resources become overburdened when clusters reach capacity requiring move operations
Solution Approach 1:
The patent implements dynamic cluster allocation where the size of TCAM clusters is not fixed but can grow and shrink based on the number of routing records they contain. Each cluster has a variable size that adapts to the actual number of records allocated to it, allowing the system to optimize between insertion simplicity and CPU resource utilization by adjusting cluster boundaries as needed.
2Reliability
If entries are kept sorted in TCAM during routing updates, then longest prefix matching is guaranteed, but routing update time increases significantly
Solution Approach 1:
The patent segments the TCAM into multiple clusters, each handling a specific range of prefix lengths. This segmentation allows routing updates to be performed within individual clusters rather than requiring global re-sorting of all TCAM entries, thus maintaining the longest prefix matching guarantee while significantly reducing update time by isolating sort operations to smaller, manageable segments.
3Ease of manufacture
If a default block size is selected for TCAM clusters, then allocation becomes simpler, but it may not suit different market applications with different IP route distribution requirements
Solution Approach 1:
The system dynamically determines cluster sizes based on the actual distribution of routing records rather than using a static default block size. The cluster allocation is adapted to match the specific IP route distribution requirements of different applications and markets, allowing the same TCAM structure to efficiently serve diverse routing scenarios without requiring manual reconfiguration.
Data Source
AI summary
Embodiments of the invention are directed to a TCAM for longest prefix matching in a routing system. The TCAM comprises a plurality of records of which a portion are configured into one or more address clusters each such cluster corresponding to a respective IP address prefix length and another portion of which are configured into a free cluster not corresponding to any IP address prefix length.


