Posted On: Jan 6, 2022
Amazon OpenSearch Service (successor to Amazon Elasticsearch Service) now offers machine learning based anomaly detection for historical data to identify trends, patterns, and seasonality in the past data. Anomaly detection for historical data enables customers to derive valuable insights from past data, and take appropriate actions to improve the overall efficiency of their applications.
Anomaly detection for historical data uses the proven, light-weight, and domain-agnostic Random Cut Forests (RCF) algorithm to detect outliers in the data, which makes it a great choice to detect anomalies in a wide range of applications. The anomaly detection feature also introduces a new unified flow in the OpenSearch Dashboards where you configure the anomaly detector just once and use it for both real-time and historical analysis. Amazon OpenSearch Service’s anomaly detection is designed to be simple and intuitive for all users, not just machine learning experts, enabling them to use the results to optimize their applications for the future.
Anomaly detection for historical data feature is powered by OpenSearch, an Apache 2.0-licensed distribution of Elasticsearch and is available on all domains running OpenSearch 1.1 or greater. To learn more about OpenSearch and its anomaly detection feature, visit the project website, and see the documentation.
Anomaly detection for historical data in Amazon OpenSearch Service is now available in Amazon Web Services China (Beijing) region, operated by Sinnet and Amazon Web Services China (Ningxia) region, operated by NWCD.