Network Interface Lookup Mechanism Based on Link Layer Semantics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network systems face challenges in scaling network workload due to loose coupling of load balancing elements, which limits collaboration between computer systems and results in inefficiencies in resource management and communication overhead, particularly in large multiprocessing systems with multiple partitions.
Innovation Solution
A network system that parallelizes network interfaces and processing, enabling flexible association of workload quanta with processing resources based on stateful or stateless associations, using a hierarchical classification scheme and optimistic matching to maximize hash table utilization, while supporting separate interfaces and addressing mutex contention overheads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional single-threaded load balancing is used, then system complexity is low, but network workload scaling capability is limited
Solution Approach 1:
The patent divides the network interface and processing into multiple partitions, with each partition handling specific workloads independently. This segmentation enables parallel processing of network packets across multiple threads and partitions, significantly improving scalability while maintaining manageable complexity through modular design
Solution Approach 2:
The patent introduces a hierarchical dimension to load balancing by implementing multi-layer classification (L2, L3, L4 layers) and multi-level partitioning. This dimensional approach allows workloads to be distributed across multiple threads, partitions, and priority levels simultaneously, enabling sophisticated scaling without linearly increasing system complexity
2Ease of operation
If loose coupling of load balancing elements is used, then ease of operation is improved, but collaboration between computer systems deteriorates
Solution Approach 1:
The patent implements nested classification where L2 classification results feed into L3 classification, which in turn feed into L4 classification. This nested structure enables tight collaboration between different classification layers while maintaining operational flexibility at each level, resolving the contradiction between loose coupling and system collaboration
3Manufacturing precision
If hierarchical classification is implemented, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent segments packet classification into distinct hierarchical layers (L2 MAC/VLAN, L3 IP, L4 TCP/UDP), with each layer handling specific classification tasks. This segmentation achieves high classification precision by examining packets at multiple protocol layers while managing complexity through clear separation of classification responsibilities at each layer
Solution Approach 2:
The patent performs preliminary classification at each hierarchical layer before proceeding to the next layer. By pre-classifying packets at L2, then L3, then L4 in sequence, the system achieves high classification accuracy without requiring all classification logic to operate simultaneously, thereby managing complexity through temporal separation of classification tasks
Data Source
AI summary
A method of performing a lookup within a network interface unit which includes providing a plurality of memory access channels, performing a multilayer lookup operation on a packet, and refining a selection of one of the plurality of memory access channels based upon the multilayer lookup operation is disclosed.