Statistics for Engineers

Friday, 2017, September 1 - 09:0012:00

Heinrich Hartmann, Circonus

Abstract: 

Statistics is the art of extracting information from data. In this workshop, we will visit the statistical methods that are relevant for operating modern IT infrastructures. Containerized cloud architectures are incredibly difficult monitoring targets. Creating probabilistic models of the behaviors of these systems, that can be used for reliable predictions is a very difficult task. In fact, it's so difficult that I don't think anyone has done that, yet. We will certainly not try to here.

Instead, we will take a different path in this workshop, and talk about statistical methods that are known to work and provided value for your daily job as a SRE. In this workshop you will learn:

  • How to measure the quality of APIs you provide and consume.
  • How to interpret the telemetry data that is emitted from the systems you are running.
  • How to aggregate metrics from single nodes to service-level views.

Topics we will cover in depths include: data visualisation, averages, percentiles, histograms, regressions, robustness and mergeability. We will cover the material from a theoretical and a practical perspective. Bring pen and paper as well as your laptop!

Course material can be found here.

BibTeX
@conference {205556,
author = {Heinrich Hartmann},
title = {Statistics for Engineers},
year = {2017},
address = {Dublin},
publisher = {{USENIX} Association},
}