Grok

神交是專為管理雲服務流異常檢測服務。

Grok is a streaming anomaly detection service built for managing public and private cloud services. Grok offers intelligent monitoring of applications and their infrastructure, resulting in superior insights into the health of critical IT systems so you can stay ahead of problems. Unlike other tools that rely on static thresholds or simple statistics, Grok learns complex patterns in your IT environment, and then identifies unusual behavior in those systems. As software, topology, and usage patterns change, Grok continuously learns, eliminating the need for frequent resetting of thresholds.

Highlights of capabilities include --

• Monitoring for public cloud providers like Amazon Web Services (AWS)

• Logical grouping of service analytics through Grok Autostacks

• Flexible Custom Metric interface allows users to securely stream data to Grok for visualization and anomaly detection

• One way integration of Grok with DataDog via the Grok CLI

• Automatic Instance Selection functionality

• Data visualization of Grok's calculated "anomaly likelihood" for monitored instances and metrics on three time-view Tabs: Hour, Day or Week, each sorted by name or degree of anomaly.