The Research Of Multilevel Security Task Scheduling Algorithm Based On Heterogeneous Multi-core Processors

Cloud storage represents the future trend of the storage services.But with the raise of the number of users,security issues are prominent increasingly.Data security and node reliability are two primary factors determine whether users choose cloud storage,and are two key factors hinder the development of cloud storage.Therefore,it is necessary to study the security of data and the reliability evaluation of nodes in cloud storage environment.Regarding the issue above,this thesis proposes a multi-level security task scheduling algorithm based on heterogeneous multi-core processor,which includes two parts: a multi-level security key mapping scheme and a task scheduling algorithm based on node reliability.The main innovations and main work of this thesis are as follows:(1)Firstly,the development trend of multi-core processors is introduced,and this thesis focuses on the architecture of homogeneous multi-core and heterogeneous multi-core processors.Abstract and build a heterogeneous multi-core processor,which mainly includes data storage node,task scheduling node and cryptographic algorithm processing node,encrypts the data based on the client in the cloud storage application scenario.(2)A task scheduling scheme based on multi-level security of data is proposed.In order to ensure data security,the mapping relationship between the file security level and the key strength is established based on the idea of multi-level security.And scheduling data to the appropriate cryptographic algorithm node to realize multi-level security encryption.(3)A task scheduling algorithm based on node reliability is proposed.In order to guarantee the reliability of cloud storage service,designs a node reliability evaluation model,which considers the subjective satisfaction of users and the objective reliability of behavior of nodes,and evaluates the reliability of nodes.The task is scheduled to a reliable storage node that matches the level of ciphertext to achieve the reliability of data storage.In summary,this thesis takes the data security and storage reliability in the cloud storage environment as the main research goal.The multi-level security key mapping scheme solves the problem that uncontrollable of data in the cloud and meets differentiated safety requirements of users.The proposed task scheduling algorithm based on node reliability guarantees the cloud storage reliability and tasks are scheduled by priority.Finally,Cloud Sim is extended to realize the proposed multi-level security task scheduling algorithm,RR scheduling algorithm and greedy algorithm.Compare and analysis of the above algorithm from several aspects,the experimental results show that the task scheduling algorithm proposed in this thesis can be used to allocate tasks more effectively and effectively,and it is more suitable for cloud storage application scenarios with diverse task requirements.