Content Router Lookup Using Hierarchical Referencing Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content routing mechanisms in communication networks are inefficient due to the complexity of managing large numbers of content names and the need for centralized architectures, leading to high latency and increased costs associated with dedicated hardware for forwarding information base management.
Innovation Solution
A method for a content router that initializes a referencing array with signatures of content names, allowing for efficient lookup and forwarding by identifying matching content names through a reduced fraction of the routing table, using hash functions and Bloom filters to manage large forward information bases with low latency and high throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional centralized content routing is used, then content delivery is possible, but routing complexity and latency increase with large numbers of content names
Solution Approach 1:
The patent segments the large routing table into multiple smaller routing tables organized in a hierarchical structure. Content names are divided into prefixes and suffixes, with the prefix portion used to index into a first-level routing table that points to second-level routing tables containing suffix information. This segmentation reduces the size of individual routing tables and enables parallel lookup operations, thereby reducing routing latency while managing complexity.
Solution Approach 2:
The patent introduces a hierarchical dimension to the routing table structure, transforming a flat, single-dimensional routing table into a multi-dimensional hierarchical structure. This involves creating multiple levels of routing tables where the first level contains prefix information and the second level contains suffix information, allowing lookups to proceed through dimensions rather than searching a single large table, thus reducing latency.
2Productivity
If dedicated hardware like FPGA or ASIC is used for FIB management, then lookup performance improves, but device cost increases
Solution Approach 1:
The patent designs a hierarchical routing table structure that can be implemented on general-purpose processors, making the solution universal and adaptable to different hardware platforms without requiring specialized FPGA or ASIC implementations. The structured organization enables efficient software-based lookups that achieve high performance on commodity hardware, reducing device cost while maintaining productivity.
Solution Approach 2:
The patent performs preliminary organization of routing information into a hierarchical structure during system initialization or configuration phases. Content names are pre-segmented into prefixes and suffixes, and routing tables are pre-organized in the hierarchical structure before actual content delivery operations begin. This preliminary action enables fast lookups during runtime without requiring complex real-time processing hardware.
3Device complexity
If content is centralized on specific servers, then content storage is simplified, but server overload occurs with frequently requested content
Solution Approach 1:
The patent segments the content delivery function across multiple servers by organizing routing information hierarchically. Different prefix-suffix combinations route to different servers, distributing the load for frequently requested content across multiple content delivery servers rather than concentrating all traffic on single centralized servers. This maintains storage simplicity while improving server capacity utilization.
Data Source
Figure 1~2
Figure 3a~4b
Figure 4c~5
AI summary
A method for a content router (CR1) in a network (NW) for forwarding a request (R) comprising a requested content name (RN) being a set of ordered components (COi), said content router (CR1) comprising a routing table (RT) and a first plurality of content names (Ni) and a plurality of forwarding interfaces (Fi), said routing table (RT) associating each content name (Ni) from the first plurality of content names (Ni) to at least one forwarding interface (Fi) from the plurality of forwarding interfaces (Fi), said method comprising the following initialization step performed by the content router (CR1): - initializing (601) a referencing array (RA) with signatures (Si) of the first plurality of content names (Ni), each signature (Si) being a representation of at least one content name (Ni), and said method comprising the following steps performed in real-time by the content router (CR1): - receiving (603) the request (R), - identifying (605) a first block (B1) from the referencing array (RA) according to a first signature (SG1) of a first subset (SS1) of ordered components (COi) from the requested content name (RN), - looking up (607) in the first block (B1) for a second signature (SG2i) of a second subset (SS2i) of ordered components (COi) from requested content name (RN), then, - if the second signature (SG2i) of the second subset (SS2i) is referenced (608) in the first block (B1), checking (609) if said second subset (SS2i) is matching a content name (Ni) from the routing table (RT), then, - if said second subset (SS2i) is matching (610) a content name (RNi) from the routing table (RT), forwarding (611) the request (R) through the forwarding interface (Fi) being associated with said matching content name (Ni).