Segmented Shared Caching for Virtual Network Routers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtualized ICN networks use dedicated caching approaches that result in random distribution of data objects across physical devices, leading to increased latency and reduced cache hit rates, despite the desire for larger local caches to improve efficiency.
Innovation Solution
Implementing a shared virtual cache that is shared by multiple virtual ICN routers, allowing for a larger cache that stores a single copy of each data object, reducing duplication and improving cache hit rates without significantly increasing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If dedicated caching is used in virtualized ICN networks, then each virtual router has local cache access (low latency), but the cache size is limited and cache hit rate is reduced due to random distribution of data objects
Solution Approach 1:
The patent merges multiple dedicated caches from different virtual routers into a single shared cache structure. The shared cache is divided into segments, with each segment accessible by specific virtual routers, but all segments form a unified cache system that increases the overall cache hit rate while maintaining low-latency access through direct memory access.
2Reliability
If larger local caches are implemented to improve cache hit rate, then more data objects can be cached, but the device complexity and resource requirements increase
Solution Approach 1:
The shared cache is segmented into multiple segments, where each segment can be independently managed and accessed by specific virtual routers. This segmentation allows the system to achieve large cache capacity without requiring a single monolithic cache structure, thereby reducing complexity while maintaining high cache hit rates.
Solution Approach 2:
The shared cache structure serves multiple virtual routers simultaneously, providing universal access to cached data objects. Each virtual router can access any segment of the shared cache, eliminating the need for separate dedicated caches for each virtual router and reducing overall system complexity.
3Speed
If data objects are distributed across multiple physical devices in dedicated caching, then local access is fast, but data duplication increases and cache efficiency decreases
Solution Approach 1:
The patent combines multiple distributed cache memories into a unified shared cache, eliminating data duplication across physical devices. The shared cache maintains a single copy of each data object that can be accessed by multiple virtual routers, thereby reducing redundancy while preserving fast access speeds through direct memory access mechanisms.
Data Source
AI summary
A method by a virtual network device in a virtualized network to support shared caching in the virtualized network. The method includes receiving a request message for content, determining whether the content is available in a shared virtual cache, wherein the shared virtual cache is shared by a plurality of virtual network devices in the virtualized network, obtaining the content from an upstream network device in response to a determination that the content is not available in the shared virtual cache, storing the content in the shared virtual cache to make the content available to the plurality of virtual network devices, and sending a response message containing the content towards an originator of the request message for the content.