Hybrid TCAM Volatile Memory Rule Lookup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions for storing and processing network forwarding rules in data centers face challenges with memory usage and lookup efficiency, particularly when dealing with a large number of prefixes or sparse rules that do not share prefixes or most significant bits.
Innovation Solution
The proposed solution involves using a combination of ternary content-addressable memory (TCAM) and volatile memory to store network forwarding rules associated with nodes of a Trie data structure. TCAM is used to store rules for the lowest level nodes, while higher level nodes, including the root node, are stored in volatile memory to optimize memory usage and reduce the time required for longest prefix match (LPM) operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If TCAM is used to store all forwarding rules, then lookup speed is improved, but memory cost increases
Solution Approach 1:
The patent segments the forwarding rules into two categories: those stored in TCAM and those stored in traditional memory. The trie data structure is partitioned such that only certain nodes (e.g., lower-level nodes with more specific prefixes) are stored in TCAM, while higher-level nodes are stored in conventional memory. This segmentation allows the system to achieve fast lookup for critical routes while avoiding the prohibitive cost of storing all rules in TCAM.
2Adaptability or versatility
If more forwarding rules are stored in memory, then routing coverage is improved, but memory usage increases
Solution Approach 1:
The patent applies local quality by assigning different storage media to different parts of the forwarding rule set based on their characteristics. Rules with shorter prefixes or less frequently accessed routes are stored in conventional memory, while rules with longer prefixes or higher priority are stored in TCAM. This allows the system to optimize memory usage by placing each rule in the most appropriate storage location based on its specific properties.
3Quantity of substance
If traditional memory is used to store all rules, then memory cost is reduced, but lookup efficiency deteriorates
Solution Approach 1:
The patent implements preliminary action by pre-processing the forwarding rules into a trie data structure and pre-determining which nodes should be stored in TCAM versus conventional memory. This pre-organization allows the lookup process to efficiently navigate the trie structure and quickly access the appropriate storage location for each rule, minimizing the performance penalty of using hybrid storage.
4Quantity of substance
If TCAM capacity is increased, then rule storage capacity is improved, but device complexity increases
Solution Approach 1:
The patent introduces dynamics by making the allocation of rules between TCAM and conventional memory adaptable and configurable. The system can dynamically adjust which rules are stored in TCAM based on changing network conditions, traffic patterns, or available resources. This dynamic approach allows the device to optimize performance without requiring a fixed, overly complex hardware architecture.
Data Source
AI summary
Examples described herein relate to configuring a device to perform longest prefix match (LPM) of rules associated with nodes to identify an action to perform on a packet. The rules can be stored among a memory and ternary content-addressable memory (TCAM) based on available memory capacity of the TCAM.