Data Resilience Protection for Power Data Center Based on Automatic Identification of Key Data and Incremental Backup

In response to the challenges brought by the significant increase in data volume and the proprietary, this study proposes a technology for automatically quantifying attributes importance and mining association rules between attributes to ensure data resilience and secure data access. An attribute identification scheme for identifying key data in power data center is designed, incorporating both importance calculation and association rule building. The scheme proposes the concept of attribute importance by leveraging the knowledge the information entropy and the maximum discrete entropy. Then, K-means algorithm is employed to cluster attributes based on its importance and Apriori algorithm is employed to further mining association rules between data attributes. Once the key data is identified, the incremental backup algorithm is employed to avoid data corruption resulted from intrusion. Experimental results show that the identification accuracy of our scheme is better than that of other identification scheme, reaching 97.7%. The joint auto identification scheme has the advantages of low cost and low false alarm rate.