Method and device for storing and searching index information

Abstract

The invention discloses a method and a device for storing and searching index information. The method comprises the following steps of: sending the index information directing to a stored content to an index node having same geographical attribution with the stored content in a convergence network for storage, after storing the content; and sending the index information directing to the stored content to the index node in a core network for storage. The invention also discloses a method for searching the index information, a device for storing the index information, a device for searching the index information and a communication network system. With the methods, the devices and the system provided by the invention, complete index information state of the content can be effectively obtained, so that implementation of flow balance and flow localization is benefited, efficiency of index searching is improved, network transmission delay caused by multi-hop searching of the prior technology is greatly reduced, and utilization rate of the index information on the index node in the core network is improved.

Description

technical field [0001] The invention relates to the field of communication technology, in particular to a method and device for storing and searching index information. Background technique [0002] In the existing distributed object storage system, the specific storage location of each object is stored in individual index files, and these index files are all stored in one or more index servers. However, in actual deployment, these index servers that manage index information are physically located in the same level, and each other is a flat architecture. [0003] In an actual deployed network, how to ensure efficient access to index information is one of the important problems faced by existing index servers. On the other hand, how to solve the flash crowd (instantaneous congestion) phenomenon (the index node becomes the bottleneck of the system) caused when a large number of users flood into the system to access a single index at the same time is another important problem ...

AI Technical Summary

Problems solved by technology

[0007] 1. In Coral, it can diffuse the index to improve multi-point access and backup. However, for the diffused index information, due to its special index registration mechanism, each node may only save a certain part of it, resulting in No node keeps a list of total index information, and it is impossible to effectively obtain a complete information state

[0008] 2. Coral uses multi-hops to spread index information to each node on the query path to reduce the load overhead of the root node. However, the multi-hop search method means that the delay of each query may increase with the number of hops. Increased and increased, affecting the query efficiency

[0009] 3. In Coral, when each leaf node caches certain index information, the index information on the root node will not be accessed by users, resulting in a low utilization rate of index information on the root node

[0010] 4. In Coral, although the concept of region is introduced in index multi-hop access, the index content data stored in each node is not divided by region. Such non-regional index results are not conducive to the realization of traffic balance and traffic localization

Images