Research And Realization Of Security Audit Technology Based On Cloud Computing Service Platform

With the continuous expansion of the network in recent years, people are more inclined to deal with the daily business using network. As the result, network provides a platform for information exchange and storage. Cloud computing technology emerged with the increase of user requirements. Many manufacturers have launched service platforms based on cloud computing technology, and relying on its versatility and low cost, the platforms are widely used. However, security issues of cloud platforms are gradually exposed. The cloud computing service providers concern more about whether users have abnormal behavior, and the audit of abnormal behavior is the most important part of the services assurance.This paper firstly introduces the cloud platform security audit-related technology comprehensive and in-depth, including learning traditional security audit standards and model, analyzing the insufficient of traditional security audit and the differences between cloud environment and traditional network environment, Establishing the theory basis for designing and realization of the audit system. On this basis, we propose a security audit model for cloud services platform. This model has the following characteristics:(1) Collect audit information based on cloud platform agent. In the information-gathering phase of the audit, through the audit agent deployed in cloud platform, the system completes the collecting and formatting of information under the cloud computing environment, and then save the data to the storage server, then conduct audit in real time. It improves the performance of audit information collected through the application of distributed cloud platform agents. (2) Security storage based on symmetric key encryption. In the process of storing audit information, by encrypting and using secure cloud storage key management system, and the use of symmetric key ciphertext retrieval, cloud service providers can prevent internal theft and tampering with data, and ensure the confidentiality of information in half-trusted cloud environment.(3) Audit analyzing that combining real-time audit and post-audit. Cloud Platform Agents do real-time audit under the rules that sent to the local rule library, audit analysis module does post-audit in accordance with the rule base. Through a combination of the two audits, it can ensure the effectiveness of the audit analysis.Meanwhile, the paper improves the link step of Apriori algorithm that used in the cloud platform post-audit, reducing the number of invalid connection and comparison, reducing the system I/O load and improve the efficiency of the algorithm.On the basis of cloud computing services platform security audit model, Beijing Industrial cloud computing platform security projects includes the main systems of audit information gathering, audit information security storage and audit information analysis. The system provides user-friendly interface and high usability. After testing, the functionality and performance of the system has achieved the expected goals.