General info
Distributed computing
systems have been more and more popular and widely used as a new way of data processing. However, in order to achieve a reliable and efficient operation of the distributed environment, it is important for providers to detect and deal with system anomalies in time.Two novel anomaly detection approaches are introduced as well as compared with the other ones. These introduced methods are devised based on data summarization and error prediction in comparison with previously extracted data.
Results
These introduced methods were devised based on data summarization and error prediction in comparison with previously extracted data.
These two novel techniques have less dependency on the predefinition of abnormal states, and additionally they are able to spot outliers with higher precision. Besides, in order to overcome number and complexity of metrics, these two methods utilize multiple review and digestion of metrics.
Check online version