So you’ve finally got Shinken up and running and you want to know how you can tweak it a bit. Tuning Shinken to increase performance can be necessary when you start monitoring a large number (> 10,000) of hosts and services. Here are the common optimization paths.
Planning a large scale Shinken deployments starts before installing Shinken and monitoring agents.
Hardware performance shouldn’t be an issue unless:
- you’re monitoring thousands of services
- you are writing to a metric database such as RRDtool or Graphite. Disk access will be a very important factor.
- you’re doing a lot of post-processing of performance data, etc. Your system configuration and your hardware setup are going to directly affect how your operating system performs, so they’ll affect how Shinken performs. The most common hardware optimization you can make is with your hard drives, RAID, do not update attributes for access-time/write-time.
Shinken needs quite a bit of memory which is pre-allocated by the Python processes.