TCAM Splitting for Large Network Search Keys
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large keys in network devices exceed the memory width of existing TCAMs, leading to inefficient storage and lookup processes, particularly in high-speed data networks where packet classification is required.
Innovation Solution
The method involves splitting keys into subsets, ranking fields by variability, and encoding metadata to reduce memory usage by storing parts of the rules in separate TCAM arrays, with intersections of overlapping fields added to ensure accurate matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If keys are stored in TCAM without splitting, then lookup speed is maintained, but memory width requirements exceed available TCAM capacity
Solution Approach 1:
The patent divides a large search key into multiple smaller sub-keys, each fitting within the TCAM's memory width constraints. The key is split into high-order and low-order portions, with metadata generated for the high-order portion to guide the lookup process in the second TCAM array, enabling efficient classification without requiring excessively wide memory
2Length of stationary object
If keys are split into multiple subsets, then memory width requirements are reduced, but the number of lookup operations increases
Solution Approach 1:
The patent performs preliminary processing by generating metadata from the high-order subset of the key before the actual TCAM lookup. This metadata is used to configure or guide the second TCAM array, preparing the system in advance to perform the second lookup more efficiently, thereby reducing the overall time penalty of multiple lookups
3Reliability
If all fields are stored in TCAM, then classification accuracy is maintained, but memory utilization becomes inefficient
Solution Approach 1:
The patent extracts only the essential high-order fields from the complete key to generate metadata, storing this compressed representation in the first TCAM array. The remaining low-order fields are handled in the second TCAM array using the metadata as guidance. This selective extraction maintains classification accuracy while significantly reducing the memory resources required in each individual TCAM array
Data Source
AI summary
A collection of rules comprising fields that may have wildcard values. The method includes defining first and second subsets of the fields, the second subset being exclusive of the first subset. Intersections of overlapping fields of the first subset are added to the first subset to form an augmented first subset. Metadata from the augmented first subset and the fields not selected for the first subset are combined to define second parts of the rules. Data items are classified by matching a search key to one of the first parts and one of the second parts of the rules.