Database Reference
In-Depth Information
Storm monitoring tools
The clustered setup of the likes of Storm need constant monitoring, because they are gener-
ally developed to support real-time systems wherein downtime could be of concern for
Service Level Agreement ( SLA ). A lot of tools are available on the market that could be
used to monitor the Storm cluster and to raise an alert. Some of the Storm monitoring tools
are as follows:
Nagios : This is a very powerful monitoring system that can be extended to gener-
ate e-mail alerts. It can monitor various processes and system KPIs and can be
tweaked by writing custom scripts and plugins to restart certain components in the
event of a failure.
Nagios service console
In the preceding screenshot of a Storm cluster with Nagios monitoring, you can see
various processes and other system level KPIs that can be monitored such as CPU,
memory, latency, HDD usage, and so on.
Ganglia : This is another widely used open source tool that lets you set up a monit-
oring framework for Storm clusters.
As seen in the preceding screenshot, we have a lot of drill-down options; we can
see load and CPU level details as well as other system and cluster level KPIs to
capture and plot out the cluster health.
Search WWH ::




Custom Search