Network Forwarding Entry Compression via Exact Match Table Lookup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network forwarding technologies face limitations in scalability and efficiency due to the high cost and power requirements of content addressable memory (CAM) for storing forwarding tables, which restricts the size of the forwarding table and leads to slower software-based lookups.
Innovation Solution
The method involves compressing destination addresses to reduce the size of forwarding entries, allowing for the use of exact match and longest prefix match tables, enabling efficient forwarding decisions while minimizing the need for expensive hardware memory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If forwarding table is stored in CAM for fast hardware lookup, then lookup speed is improved, but cost and power consumption increase significantly
Solution Approach 1:
The patent divides the forwarding table into two segments: a first forwarding table stored in CAM for exact match lookups, and a second forwarding table stored in conventional memory for prefix match lookups. This segmentation allows the system to use fast hardware lookup only when necessary (exact matches) while using slower but cheaper software-based lookup for other cases, thereby reducing overall power consumption and cost while maintaining acceptable lookup performance.
2Adaptability or versatility
If forwarding table size is increased to improve forwarding capacity, then network scalability is improved, but CAM cost and power requirements increase
Solution Approach 1:
The patent segments the forwarding table storage into CAM and conventional memory, allowing the system to scale forwarding capacity by adding entries to the second table in conventional memory without proportionally increasing CAM power consumption. The first table handles exact matches with limited capacity, while the second table provides scalable prefix match capacity using cheaper memory technology.
Solution Approach 2:
The patent changes the storage medium parameter from exclusively CAM to a hybrid of CAM and conventional memory. This parameter change allows the system to maintain forwarding capacity while significantly reducing power consumption for large forwarding tables, as conventional memory consumes much less power than CAM.
3Use of energy by stationary object
If forwarding table is stored in main memory for software processing, then cost is reduced, but lookup speed decreases
Solution Approach 1:
The patent segments the lookup process into two paths: exact match lookups that utilize fast CAM hardware acceleration, and prefix match lookups that use conventional memory with software processing. This segmentation ensures that time-critical exact matches benefit from hardware speed while less time-sensitive prefix matches use the energy-efficient software path.
Solution Approach 2:
The patent introduces a compression module as an intermediary that compresses destination addresses before lookup. This compression reduces the size of forwarding entries in conventional memory, improving cache hit rates and reducing memory access time, thereby bridging the speed gap between CAM and conventional memory without increasing power consumption.
Data Source
AI summary
A method and apparatus of a device that determines a match for a compressed address using an exact match table of a network element is described. In an exemplary embodiment, the network element receives a data packet that has a destination address. The network element further compresses the destination address to give a compressed address. In addition, the network element performs an address lookup using the compressed address in an exact match table. Furthermore, a match in the address lookup indicates a transmitting interface of the network element. The network element forwards the data packet using the transmitting interface.