Packet Classification Fingerprint Database Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current packet classification methods in data networks, such as those using Access Control Lists (ACLs), face challenges with large rule sets and high memory requirements, particularly in terms of power, cost, and space constraints, especially with increasing packet 'intelligence' and the need for sophisticated filtering and routing in modern networks.
Innovation Solution
The method involves representing packet classification rules in a compressed form using rule fingerprints, which are smaller and can be stored in limited on-chip memory, allowing for efficient matching operations and minimizing accesses to the full rules database by first using a fingerprint database to identify possible matches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If the full rules database is stored in on-chip memory for fast packet classification, then packet classification speed is improved, but the power consumption, cost, and space requirements increase significantly
Solution Approach 1:
The patent divides the rules database into two segments: a compressed fingerprint representation stored in on-chip memory for fast lookup, and the full rules stored in off-chip memory. This segmentation allows the system to maintain high-speed classification while reducing on-chip memory requirements and power consumption.
Solution Approach 2:
The patent creates a compressed copy (fingerprint) of the rules database that retains essential matching information while occupying minimal space in on-chip memory. This copy enables fast packet classification without requiring the full rules database to reside in high-power on-chip memory.
2Speed
If the full rules database is stored in on-chip memory for fast packet classification, then packet classification speed is improved, but the device cost and space requirements increase significantly
Solution Approach 1:
The patent segments the rules database storage between on-chip and off-chip memory, placing only the compressed fingerprint in on-chip memory. This reduces the area occupied by on-chip memory while maintaining fast classification capability through the compact fingerprint representation.
Solution Approach 2:
The patent transforms the rules database into a compressed fingerprint representation, changing the parameter of data density. This compression dramatically reduces the space required in on-chip memory while preserving the essential information needed for packet classification.
3Volume of stationary object
If algorithmic solutions like MRFC are used for packet classification, then memory capacity is reduced, but the number of external memory accesses increases
Solution Approach 1:
The patent performs preliminary compression of the rules database into fingerprints before packet classification. This pre-processing enables fast on-chip matching without requiring multiple external memory accesses, thus reducing both memory capacity requirements and access time.
Solution Approach 2:
The patent introduces fingerprints as an intermediary between packets and the full rules database. This intermediary enables fast matching in on-chip memory, reducing the need for slow external memory accesses while maintaining accurate packet classification.
4Measurement precision
If TCAM is used for rapid table lookups in ACL processing, then packet classification accuracy is improved, but the power, cost, and space constraints are exceeded
Solution Approach 1:
The patent creates a compressed copy (fingerprint) of the ACL rules that can be stored in standard on-chip memory instead of power-hungry TCAM. This copy maintains classification accuracy while dramatically reducing power consumption and cost.
Solution Approach 2:
The patent changes the representation parameter of ACL rules from full-length entries to compressed fingerprints. This parameter change enables accurate packet classification using standard memory technologies instead of expensive, power-intensive TCAM.
Data Source
AI summary
In one aspect, the present invention reduces the amount of low-latency memory needed for rules-based packet classification by representing a packet classification rules database in compressed form. A packet processing rules database, e.g., an ACL database comprising multiple ACEs, is preprocessed to obtain corresponding rule fingerprints. These rule fingerprints are much smaller than the rules and are easily accommodated in on-chip or other low-latency memory that is generally available to the classification engine in limited amounts. The rules database in turn can be stored in off-chip or other higher-latency memory, as initial matching operations involve only the packet key of the subject packet and the fingerprint database. The rules database is accessed for full packet classification only if a tentative match is found between the packet key and an entry in the fingerprint database. Thus, the present invention also advantageously minimizes accesses to the rules database.