A multi-attribute probabilistic caching method in named data network

Abstract

The invention discloses a multi-attribute probabilistic caching method in a named data network, comprising: a routing node judges the attribute of a Data packet; if the attribute of the Data packet is an unshareable attribute, the data carried in the Data packet is not cached; if the Data packet If the attribute is a shared attribute, then judge whether to cache the data carried in the Data packet according to the node cache probability; among them, P=P(v)+P(u), P(v)=β / (ɑ+β), P( u)=u / (u+θ); where P is the node cache probability, P(v) is the node centrality probability, P(u) is the node distance probability; β is all the Interest entries in the routing node PIT The sum of the number of ports, α is the sum of the total number of routing nodes in the NDN network; u is the number of hops that the Data packet passes through the routing nodes, and θ is a positive integer greater than 0. The present invention effectively reduces the redundant storage of the same data by different routing nodes in the NDN network, saves the use of routing node caches, reduces network delay, and improves the performance of the NDN network.

Description

technical field [0001] The invention relates to the technical field of caching strategies, in particular to a multi-attribute probabilistic caching method in a named data network. Background technique [0002] The modern Internet is a communication network on a global scale, which has become an important infrastructure for global communication. The current Internet architecture is designed as an end-to-end connected underlying communication framework for data transmission. With the increase in the number of users and the amount of communication data, this terminal-centric, location-based communication framework has problems such as poor mobility, poor reliability, and inability to adapt to changes in network upper-layer applications. As the contradiction between the existing Internet architecture and the ever-changing user needs is increasingly intensified, the research and development of the future Internet architecture has become a hot issue of common concern in both acad...

AI Technical Summary

Problems solved by technology

Although the default LCE caching strategy of the NDN network is simple, this strategy will cache the data carried by the Data packet at each routing node, causing the same data to be redundantly cached in different routing nodes, wasting cache resources

This caching strategy will also cause the cached data in the routing node to accelerate the speed of cache replacement, so that important data cannot be stored in the routing node for a long time

The LCD and MCD algorithms only cache the data carried by the Data packet in the next routing node after the Interest hits Data in the local node. The advantage of this design is that compared with LCE, it can reduce the cache of the same redundant data in the routing node, but in reality In the production environment, multiple users usually request to obtain the same data resource. This caching strategy cannot cache the data resource in the routing node closest to the user in a timely manner, resulting in the need to obtain the same data from a farther data source when requesting again. Get data, increase network delay

The Prob caching strategy sets a fixed caching probability for data at each routing node. This caching strategy is relatively simple and does not consider factors such as distinguishing the attributes of the data produced by the producer; for example, shared resources such as videos, pictures, and documents need to be cached. But non-shared resources like chat messages don't need caching

Images