Research on parallel detection of heterogeneous cloud resources with multiple anomalies in cross-type database

In the current multi anomaly parallel detection of heterogeneous cloud resource cross database, the processing of data characteristics is insufficient, resulting in low accuracy of detection results. Therefore, this paper proposes the research on multi anomaly parallel detection of heterogeneous cloud resource cross database. Preprocess heterogeneous cloud resource cross data. Select the local features of the data for depth feature synthesis. Analyze the deviation relationship of core indicators of anomaly detection. Parallel detection of database multiple abnormal behaviors. The experiment shows that the average accuracy rate of the detection results of this research method is 0.987927, which has been greatly improved and has high practical application value.