Research Of The Key Technologies In Cloud Security Immune System

Along with the coming of the era of big data, great changes of the data volume, data complexity and data processing have been taking place. Cloud computing is regarded as the third information revolution after the PC and the Internet. It embodies the thought of "network is the computer" and can fuse huge amounts of software and hardware resources effectively to form a huge shared computer resource pool and provide convenient and rapid on-demand service for users, which meets the actual demand of users better under the background of big data.With the further development of cloud computing, cloud data security has become the key problem. Cloud computing are facing the challenges of how to make sure the security of user data in the environment of mass nodes and needs of information processing. The traditional technologies of cloud security can not satisfy the perfect time and space efficiency in cloud data processing. Its dynamics and intelligence is poor. Further more, it occupies more of the bandwidth of the service.In order to protect the safety of data operation under the cloud environment better, setting the Human Immune System (HIS) operating mechanism as a theoretical basis, using its features for reference, based on artificial immunization and Computer Immune System (CIS), combined with the condition of the cloud environment, this thesis comes up with Cloud Security Immune System (CSIS) based on the principle of Human Immune System. It designs operating mechanism and eight related algorithm of CSIS, analyzes its characteristics. Experiments show that the CSIS has higher recognition rate of the abnormal feature data and the self-set feature data, which can well protect the security of cloud data.Due to improve the performance of CSIS, the thesis focuses on the representation of self-set feature data and storage. Firstly, it proposes a Bloom Filter Self-set storage model (BFSS), which can reduce the storage space of self-set effectively and save query time. Secondly, in order to support the delete operation of the feature data, it put forwards a Counting Bloom Filter Self-set storage model (CTBFSS). Finally, to reduce the storage space of self-set and reduce querying time further, it raises Compressed Bloom Filter Self-set storage model (CPBFSS).Experiments demonstrate that BFSS, CTBFSS and CPBFSS can reduce the time cost between customer and service effectively on the premise of higher recognition rate, speed up the search procedure and improve the overall performance of the system.