Home / Educational Content / Database & Technology / Analytics Using Feature Selection for Anomaly Detection

Analytics Using Feature Selection for Anomaly Detection

Presented at INSYNC 21

Session ID: 100690

Experience an innovative approach which relies on big data and advanced analytical techniques to detect anomalies in performance metrics in order to improve Oracle DB performance. The approach represents a step-change paradigm shift away from traditional methods which tend to rely on a few hand-picked/favorite metrics or wading through a voluminous AWR report. With this innovative process, you can draw on all available performance data to help you draw impactful, focused performance improvement conclusions.

The process targets the most relevant metrics for the problem interval which results in event focused and actionable intelligence on the performance issue, and insight into possible solutions. This method improves on the typical/traditional approaches which monitor a standard set of hard coded metrics; which may cause you to miss important system behavior or configuration that is relevant to the root cause of the performance problem. The speaker will share his process and code.

Watch the full-length video

Premium Content: access is limited to Quest Corporate and Professional members.

Membership has its perks. Get unlimited access to the latest Oracle updates, event session replays, strategic content centers and special members-only programming, plus big discounts on conference registration, with a Quest Corporate or Professional membership. Quest is where you learn.

Analytics Using Feature Selection for Anomaly Detection