Using Unsupervised ML to Detect Anomalies in Machine Data by Dan Turchin

preview_player
Показать описание
Title: Using Unsupervised ML to Detect Anomalies in Machine Data: How InsightFinder predicts and prevents IT outages with automated root cause analysis

Speaker: Dan Turchin, CEO of InsightFinder (Twitter @dturchin)

In this discussion, InsightFinder CEO Dan Turchin will share where we’ve been as an IT Ops community, where we are, and where we’re headed. He’ll discuss the unique technology breakthroughs achieved by InsightFinder and demo the future of application observability.

For a decade, IT Operations teams have struggled to manage increasing data volumes from metrics, logs, traces, and change events. That, combined with new architectures like microservices, API-driven everything, serverless computing, and auto-scaling across multiple clouds make systems management an unreasonable task for humans without help from machine intelligence.

InsightFinder is the first AI platform designed to detect anomalies in machine data at scale. It uses unsupervised machine learning from streams of structured and unstructured data to analyze billions of rows of cross-source data daily. Unlike alternatives, InsightFinder automates the lifecycle of operational incidents by:

- Detecting anomalies across data sources
- Predicting incidents five to seven hours before they occur
- Reducing noisy alerts by up to 90%
- Automating root cause analysis and remediation
Рекомендации по теме
join shbcf.ru