7–9 November 2016: Conference & Tutorials
9–10 November 2016: Training
Amsterdam, The Netherlands

A hands-on data science crash course on web performance monitoring with Python

Bart De Vylder (CoScale)
14:00–15:30 Wednesday, 9/11/2016
Metrics/monitoring DevOps G104/105 Audience level: Intermediate
Average rating: ****.
(4.00, 6 ratings)

Prerequisite knowledge

  • Programming experience (preferably in Python)
  • Basic mathematical and statistical knowledge

Materials or downloads needed in advance

  • A laptop (We'll use an iPython notebook made available online, avoiding the need for you to install anything.)

What you'll learn

  • Gain hands-on experience with data analysis on a real-world dataset, using easily accessible tools and relatively simple yet powerful techniques


Data science is a hot topic these days. Bart De Vylder breaks through the hype to provide a practical introduction to data analysis techniques applied to web performance data and related business metrics using Python.

Python is well known as a general purpose programming language. Moreover, the gap between interactive analysis and writing production-ready analytics code in Python is relatively small. Bart uses the iPython scientific ecosystem with NumPy, SciPy, and scikit-learn to make data analytics very approachable.

Drawing on a real-world anonymized dataset of monitoring data of backend server metrics, real-user monitoring data, and business metrics originating from an ecommerce website, Bart guides you through some techniques for visualizing (large) datasets, finding correlations between metrics (e.g., page load time versus conversion rate), applying machine-learning techniques to build a model of the data (e.g., which requests cause the most load on a server), anomaly detection, and forecasting of data. Bart ends with a challenging problem on the given dataset using one of the discussed techniques—with a nice prize for the attendee with the best solution.

Photo of Bart De Vylder

Bart De Vylder


Bart De Vylder is a data scientist at CoScale. Previously, Bart was active in software engineering and architecture, with a focus on distributed systems. His interests lie in machine learning and building reliable, scalable data processing systems. Bart holds a PhD in artificial intelligence from the Free University of Brussels.