DevOps Pro Moscow 2022

24 - 26 мая


 Москва и онлайн

30 мая – 3 июня



Alexander Acker

Должность:  Post-doc Researcher

Компания:  TU Berlin

Страна:  Germany


Dr. Alexander Acker is a postdoc researcher at the group on distributed and operating systems of the Technical University of Berlin and CEO of the startup His work is focused on using artificial intelligence and machine learning methods for the autonomous operation of complex IT systems to improve their dependability. A. Acker has six years of professional experience including both, industry and academia. He was involved in several industrial projects with global cloud, telco, and manufacturing companies, developing solutions that improve the dependability of different IT systems. He published papers on top-ranked international conferences such as HICSS, ICSOC, ICDM, IEEE BigData, and ECML-PKDD.


AI-Enabled Log Analytics for Proactive Troubleshooting

Artificial Intelligence for IT Operations (AIOps) combines big data and machine learning to replace a broad range of IT operations tasks including availability, performance, and monitoring of services. In this talk, we will focus on AI-empowered analytics of log lines to detect suitable starting points for debugging and troubleshooting, as these methods are highly relevant in practical applications and help to bridge the gap between devs and ops. Also, this talk will present the basic idea, business reasoning, and potential behind AI-empowered log analytics and introduce a typical structure of the corresponding platforms. An overview of sample tools available for the development and operation cycle will be complemented by a description of a typical pipeline for the collection of log lines, analysis, remediation as well as incident visualization. A demonstration of a corresponding web service implementing the AI-based log analytics together with features for detection of flow anomalies, for quality and variable analysis as well as for continuous verification will conclude the talk.

Ключевые слова

Deep Learning
Log Analytics
Anomaly Detection

« Hазад