enterprise getting started

Installing FnordMetric Enterprise is straightforward. Download the latest release

[here](/documentation/downloads) and run the jarfile with this command:

    java -jar FnordMetric-Enterprise-v1.2.7.jar --tcp 8922 --udp 8922 --websocket 8080

    java -jar FnordMetric-Enterprise-v1.2.7.jar --tcp 8922 --websocket 8080 --admin 8081

This will start FnordMetric, listen on UDP and TCP port 8922 and start a WebSocket

server in port 8080.

This will start FnordMetric, listen on TCP port 8922, start a WebSocket server on port

8080 and start the admin interface on port 8081.

### Semantics

There are three basic operations: `add\_sample`, `value\_at` and `values\_in`.

`add\_sample` adds a sample to a metric, `value\_at` retrieves the measured value at

a specified time and `values\_in` retrieves all aggregated measured values in a specified

time interval.

A metrics type and interval are implicitly specified by the metrics key; all keys have to

end with "$type-$interval".

Each metric is addressed by a unique key. The metrics type and interval are implicitly specified

by its key; all keys have to end with a "$type-$interval" pattern.

For example: if you want a metric `response\_time` to record the average/mean of all sampled

values in an aggregation interval of 60 seconds, use the key `response\_time-mean-60`. For a

metric `total\_clicks` that sums up all measuements in one-hour intervals, you could use

`total_clicks.sum-3600`

values in an aggregation interval of 60 seconds, you'd use the key `response\_time-mean-60` (type: mean,

interval: 60s). For a metric `total\_clicks` that sums up all measuements in one-hour intervals, you would use

`total_clicks.sum-3600` (type: sum, interval: 3600s)

You can find a list of all metric types and the [full API Reference here](/documentation/fnordmetric_api_reference/)

### Getting Started

...

As an example, we will report the average http response time of a fictional application "myapp" to

FnordMetric. They key we will use for this is `myapp.response\_time-mean-10` which will create a

metric "myapp.response_time" that records the mean of all sampled values in a 10 second interval.

To add the first value to this metric we use netcat from the command line. This opens up a TCP

connection to localhost on port 8922 and sends one line to add the value 42 to the metric. In our

example this means we record the response time of one HTTP request that took 42ms.

FnordMetric Enterprise will automatically create the metric if it doesnt exist yet.

    $ echo "SAMPLE myapp.response_time-mean-10" | nc localhost 9222

    OK

There are also [other ways to connect to FnordMetric Enterprise](/documentation/fnordmetric_api_reference/)

(e.g. WebSockets, UDP, HTTP).

You should now be able to navigate to the admin interface on http://localhost:8081/ in your

browser and see something like this:

<img src="/img/fnordmetric_ui_example_screen.png" width="630" class="shadow" />

<br />

We can retrieve the measured data in a similar way. This will e.g. retrieve the measured mean

response times for the last hour ([see a full list of commands here](/documentation/fnordmetric_api_reference/))

    $ echo "VALUESIN myapp.response_time-mean-10 -1h now" | nc localhost 9222

    1360804571:4233.52 1360804581:4312.36 1360804591:6323.12

You now have a running FnordMetric application. There is a lot more you can do: Checkout out

the [full API Reference here](/documentation/fnordmetric_api_reference/) or [read more

about FnordMetric UI](/documentation/ui_index), a JavaScript / HTML5 library that you can use

to plug in the data into any webpage within seconds.