Nov 07, 2018
 Percona Live Open Source Database Conference Europe 2018
  Kay Agahd

How to visually spot and analyze slow MongoDB operations

Do you ever wonder why your MongoDB databases seem to be much slower than expected? Do you experience periodical hot spots and want to know the culprit(s)? Do you want to know which database operations are the most expensive and the ones most meaningful to improve?

In this talk you will quickly understand the limits of MongoDB's built-in database profiler, especially in sharded environments. I will present an open source tool we developed to solve these issues. We created a tool that collects, stores and helps to visually analyze slow MongoDB operations. By using the tool you will get a better picture of your MongoDB clusters without the need of a deeper MongoDB expertise. The tool runs database commands simultaneously against a multitude of databases or mongod's and with just a few clicks you can easily intervene and spot performance issues.



About the Author

Kay Agahd

Experienced Software and Database Engineer with close to two decades experience in the internet industry. Worked for 10 years as a Software Developer in Paris/France and since 2009 in Berlin/Germany. Specialized in using MongoDB including sharding in production since 2011 (since version 1.6). Skilled in Java, relational and NoSQL-databases, automation, profiling, monitoring, performance optimisations and design of large scale databases. Strong IT background with an academic degree in Computer Software Engineering from Hamburg University of Applied Sciences. MongoDB Certification both as Java developer (final grade 100%) and as DBA (final grade 96%).