Tuning Your DBMS Automatically with Machine Learning

Click for: original source

Dana Van Aken, Geoff Gordon, and Andy Pavlo from Carnegie Mellon University guest blog post on AWS demonstrates how academic researchers can leverage AWS Cloud Credits for Research Program to support their scientific breakthroughs.

Database management systems (DBMSs) are the most important component of any data-intensive application. They can handle large amounts of data and complex workloads. OtterTune, a new tool written in Python that’s being developed by students and researchers, can automatically find good settings for a DBMS’s configuration.

Article then goes on explaining:

  • How OtterTune works
  • The machine learning pipeline
  • Implementation
  • Evaluation with MySQL and Postgres

Also loads of charts supporting the explanation. Nice!

[Read More]

Tags machine-learning data-science database