Novel index system of network conversation package

Abstract

The invention provides a novel index system of a network conversation package. The system comprises at least one index space used for storing retrieval information of a type of information of the network conversation package, an information compression module, a retrieval information projection module, and a reading module. A plurality of storage positions are arranged in the index space, one storage position corresponds to a type of retrieval information, and the storage positions have the same initial values; the information compression module is used for performing compression operation on one or more types of information of the specified network conversation package to obtain the retrieval information corresponding to the information; the retrieval information projection module is used for projecting and mapping related information of the network conversation package to the corresponding storage positions of the index space which stores the retrieval information of the information, and the projected storage position is marked to be already projected when being projected for the first time; and the reading module is used for finding the corresponding network conversation package. According to the system, the system performance can be greatly improved.

Description

technical field [0001] The invention belongs to the field of data indexing, in particular to a novel indexing system for network session packets. Background technique [0002] The current index is basically based on the quadruple as the KEY to build the index. Under the massive data traffic, the processing performance and capacity of the HASH table cannot be satisfied. At present, the mainstream HASH processing limit is about 10 million, and the memory size occupied by the index block is not fixed, and it increases with the increase of the data volume. It is impossible to pre-allocate fixed memory, and it needs to endure frequent and invalid memory application release. Contents of the invention [0003] In order to solve the above problems, the invention provides a novel indexing system of a network session packet, and the system includes the following modules: [0004] At least one index space used to store retrieval information of certain information in the network sess...

AI Technical Summary

Problems solved by technology

[0002] The current index is basically based on the quadruple as the KEY to build the index. Under the massive data traffic, the processing performance and capacity of the HASH table cannot meet the requirements.

At present, the mainstream HASH processing limit is about 10 million, and the memory size occupied by the index block is not fixed. It increases with the increase of the data volume. It is impossible to pre-allocate fixed memory, and it needs to bear frequent and invalid memory application releases.

Images