Collect Java/JVM application metrics in production
To collect metrics in production it is important to keep the overhead low and for the statistical counters to
not introduce contention. JSR166e brings us
LongMaxUpdater which were
built just for this task.
Avaje metrics won't introduce contention or blow out your memory.
Collecting metrics in minutes
An agent is provided so that you can instrument your JVM application without
having to write code. Classes annotated with
@Service etc can be instrumented automatically.
Within minutes you can be gathering metrics across all the layers of your application.
Timers, Counters, Gauges
There are 4 main types of metrics that are supported.
- Timers - monitor timed events
- Gauges - monitor resources like memory and threads
- Counters - monitor discrete events
- Values - monitor events like 'bytes sent'