Data name search method, system and storage medium

Abstract

The invention discloses a data name search method, system and storage medium. On the basis of the existing data name search algorithm, an extra step is added. When the data name prefix is ​​stored in the hash table according to the number of components, the data name prefix The first n bits of each component are taken out as the feature prefix of the data name prefix, and the feature prefixes of the same components are counted together and stored with Bloom filter. When it is necessary to find the target data name, first obtain the characteristic data name of the target data name according to the corresponding method, and then put the characteristic data name into the characteristic prefix Bloom filter corresponding to the number of components for matching. If the matching fails, there is no need to enter the corresponding Lookup in a hash table of component numbers. The invention quickly completes the data name search and reduces the network delay.

Description

technical field [0001] The invention relates to the field of computer network transmission, in particular to a data name search method, system and storage medium. Background technique [0002] Unlike the traditional IP network, which is address-centric, NDN, as a new network architecture, shifts the focus of the network from data location to data content. That is, in a traditional IP network, users need to find the IP address of the desired content first, and then find the server location where the content is located before they can get the content; while NDN only pays attention to the content required by the user, not where it comes from, which is more in line with the user's search habits. Therefore, the transmission of NDN is not the same as that of traditional IP networks. In IP search, only the 32-bit (IPv4) or 128-bit (IPv6) IP address "01" string needs to be searched. It is much more complicated: first, each data name consists of several components, and the number of...

AI Technical Summary

Problems solved by technology

However, this method does not take advantage of the characteristics of the prefix itself, but only stores the data name prefix according to the number of components, which does not have the effect of filtering the data name prefix at all, making the query efficiency not high

Especially in the scenario where there are many times of data name matching failures, the search efficiency of this method is low, because each search will go to the hash table with a component number of 1 and the search fails. The number of searches is large, resulting in an increase in the number of memory accesses and a large network delay

[0008] All in all, the existing method will generate a large number of hash search processes in each data name search process, which makes the number of memory accesses in the data name search process more and the search efficiency is low; there are a large number of hashes in the existing data name search method. Hash search process, and the time occupied by the hash search process accounts for a large proportion of the entire search process, making the delay in the network transmission process relatively high

Images