Real-time anomaly detection over VMware performance data using storm

摘要: Anomaly detection is the identification of items or observations which deviate from an expected pattern in a dataset. This paper proposes novel real time anomaly framework for dynamic resource scheduling VMware-based cloud data center. The monitors VMware performance stream (e.g. CPU load, memory usage, etc.). Hence, continuously needs to collect and make decision without any delay. We have used Apache Storm, distributed handling making prediction Storm chosen over traditional (e.g., Hadoop MapReduce, Mahout) that good batch processing. An incremental clustering algorithm model benign characteristics incorporated our storm-based framework. During continuous incoming test stream, if finds deviated its behavior, it considers as anomaly. shown effectiveness by providing real-time complex analytic functionality data.