GitLab Commit 2019 - Brooklyn

Logging and Data analysis are intrisincally tied. From an operational perspective, backend storage must be aligned in order to centralize the data. Traditional data analysis is not fast enough for modern needs. Pushing logs to aggregators for processing is time-consuming and slows the feedback loop. There are plenty of opportunities to improve data analysis time while the data is still in motion. In this presentation, Eduardo will discuss data processing patterns and how this new approach of Stream Processing on the Edge can be used for faster data availability and smaller feedback loops in distributed environments. The result? Systems that can evolve faster and data science teams are empowered to accurately understand what is happening inside their systems.