Metrics/Analytics

[bketelsen]

Each RPC service should export statistics on request timing, average, max, min, etc.

We should be able to see timing on each component of a composite service. If there are 10 RPC calls, then we should see timing on all ten as a piece of the composite.

[bketelsen]

We all love metrics and graphs, skynet will make sure you get your daily fix. More than likely we will utilize something like graphite to log time series data regarding number of requests, which calls are being made, response times, cpu/memory/load so that you can determine the state of your skynet cluster quickly with a dashboard of metrics.

After this functionality has been added we'd like to create a nice web interface to be hit and see live statistics going on across the system at that moment and refreshed live.

__
Make it easy to add new metrics. Use something like https://github.com/peterbourgon/g2g

[erikstmartin]

I plan to implement an interface so packages can be created to register themselves similar to the sql package.

Create a metrics package to report metrics to https://tempo-db.com
Also leverage https://github.com/jondot/gosigar to report CPU/Mem/Load/Disk to doozer for load balancer, and to the metrics interface

[bketelsen]

Half completed in #49