Distributed Cache Coordination for Mobile Network Hit Rates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In mobile packet data networks, caching systems face inefficiencies due to limited transmission capacity and low cache hit rates in resource-constrained environments, where small caches serve limited populations, reducing the likelihood of content being requested by others.
Innovation Solution
A distributed cache architecture is implemented, using a cache manager to coordinate data distribution between caches, optimizing the use of transmission capacity by pre-loading content during off-peak hours and prioritizing high-priority data, effectively creating a large virtual cache from multiple small ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If multiple small caches are deployed in RBSs to serve local terminals, then transmission capacity is saved locally, but cache hit rate decreases due to limited population
Solution Approach 1:
The patent merges multiple small caches in different RBSs into a unified cache domain where caches share information about requested objects. This allows a cache serving a small local population to benefit from the collective demand of the entire cache domain, effectively increasing the population served without physically expanding individual cache capacities.
Solution Approach 2:
The patent introduces an intermediary mechanism (cache information sharing protocol) that enables caches to exchange information about requested objects. This intermediary system allows caches to serve requests from objects requested by other caches in the domain, effectively creating a virtual larger population without requiring direct physical connection to all terminals.
2Reliability
If caches store popular content to increase cache hit rate, then more transmission capacity is consumed during off-peak hours for content distribution
Solution Approach 1:
The patent implements preliminary action by distributing content to caches during off-peak hours when transmission capacity is available. Caches proactively store popular objects before peak demand occurs, ensuring high cache hit rates during peak hours without consuming transmission capacity during those critical periods.
Solution Approach 2:
The patent employs periodic action by scheduling content distribution to occur during specific off-peak time periods. This allows the system to periodically refresh and update cache contents with popular objects, maintaining high cache hit rates while utilizing transmission capacity only when the network is less congested.
3Reliability
If a large centralized cache is used to serve all terminals, then cache hit rate increases, but transmission links become bottlenecks and complexity increases
Solution Approach 1:
The patent segments the caching function across multiple distributed RBSs rather than using a single centralized cache. Each RBS maintains its own cache storage and can independently serve local requests, while the segmentation is logically unified through information sharing. This reduces network complexity and avoids creating transmission bottlenecks at a single centralized point.
Solution Approach 2:
The patent adds a new dimension to the caching architecture by introducing inter-cache communication and information sharing across the radio access network. Instead of a single-dimensional centralized or local cache model, the system creates a multi-dimensional distributed cache domain where caches at different RBSs cooperate, effectively increasing cache hit rate without centralizing physical storage.
Data Source
AI summary
A method for optimising the distribution of data objects between caches in a cache domain of a resource limited network. User requests for data objects are received at caches in the cache domain. A notification is sent from each cache at which a request received to a cache manager. The notification reports the user request and identifies the requested data object. At the cache manager, object information including the request frequency of each requested data object and the locations of the caches at which the requests were received are collated and stored, and objects for distribution within the cache domain are identified on the basis of the object information. Instructions are sent from the cache manager to the caches to distribute data objects stored in those caches between themselves. The data objects are distributed between the caches using transmission capacity of the network that would otherwise be unused.