A Study Of Hash Function And Conflict Resolution Base On Signature

With the rapid development of the Internet, cloud and big data are concerned by many companies in these years. people begin to enjoy the convenience brought by these emerging technologies, while also suffer pervasive network intrusions. Network security has become related to the vital interests of you and me another important issue. Snort virtue of its C language features, lightweight, open source and other characteristics, is used. How to better organize fingerprint Snort rule base, is the focus of developer concern. And how to find a faster network intrusions, it is the problem of interest to the user. Especially applied to the terminal, to match the speed of its data packet determines its performance advantages and disadvantages, became the key to its success.As the framework is based on hardware implementation, so the matching rate and a higher storage space requirements, the focus of this article is without prejudice to the matching rate and the percentage of cases of conflict of how to get a lower rate.To solve this problem, the work of this paper are as follows:1. The framework for space collision rate and the hash table has an extremely stringent requirements, and the required map data set is dynamic, coupled with the lack of consideration of the original hash function, this paper designed a new hash function. For different data sets, this function will be digitally statistical method for data preprocessing, while the space after the hash function maps will remain in a smaller space. Get inspiration from classic hash function, which in the design process to fully take into account the stability problem, is a good hash function.2. The framework of the original conflict deal is for a particular data set, while access to the lower collision rate, but it is at the expense of reaching the premise of the hit rate of the packet, and its method itself is used in a tentative the method, less stable. In this paper, starting from the nature of the conflict, to consider how to eliminate the root causes of conflict, and thus devised a new method of handling conflict. Meanwhile, the method using two hash structure, reducing the degree of coupling between the DFA and the fingerprint characteristics, thereby further reducing the rate of collisions.Finally, experimental results show that without affecting the hit rate, the new hash functions and conflict resolution methods can reduce the collision rate to 1.72 ‰.