Vyve Uses Augtera Networks to Prevent Incidents. Click here to learn more.

Metric Extraction From Logs

Metric Extraction Overview

Data types such as SNMP and sFlow are structured with metric extraction in mind. Unstructured log messages often focus more on human readability, though they also contain a wealth of metric data. Augtera’s LogAI can extract metrics from unstructured log messages, apply AI/ML algorithms developed for metrics, detect anomalies, eliminate noise, and notify ServiceNow, Slack, customer Automation systems and more.  

Metric Extraction From Logs

Log messages contain metrics. However, when embedded in an unstructured log message, these metrics are hard to extract, analyze and visualize with network operations tools. Augtera enables the extraction of metrics from log messages. Metric extraction can be used for any log message, regardless of where in the network the message pertains to.  

For example, metrics are extracted from raw log messages like the one in the below image and are then available for timeline visualization and AI/ML algorithms. 

Example of metric extraction from a log message, and metric visualization

Applying AI/ML Algorithms to Extracted Metrics

The extraction process is not limited to specific messages and can be leveraged with all Augtera purpose-built algorithms for networking. Many of them apply to anomaly detection on metric data. Once metric data is extracted from logs, the full power of the Network AI platform can be leveraged for anomaly detection, noise elimination, incident root identification, and notification (Slack, ServiceNow, Automation processes, … ). 


LogAI is not just another tool for storing log messages. It mines log data at the speed of streaming messages, finding rare / new messages that precede outages, detecting burst patterns, and extracting metric data so metric-oriented algorithms can be leveraged. LogAI detects and acts as logs are being streamed, it is not another store and query log solution, it is purpose-built for Network Operations. 

To learn more about LogAI: