Monitor and record time to run Athena queries

[fbertsch]

There are quite a few options to this:

1. Monitor and Alert within MC
2. Collect metrics within MC and output to a third-party service e.g. datadog
3. Collect metrics with a tool like statsd and output to a third-party service ...etc.

But with monitoring we need to alert as well.

[fbertsch]

One option: https://docs.datadoghq.com/developers/dogstatsd/

[akkomar]

FYI: we're already using DogStatsD on Databricks (might be useful for setting this up).

[wlach]

Let's try to get this done sooner rather than later, in case we run into problems.

[wlach]

On second thought, I think new relic is probably sufficient for tracking major problems, let's move this to the next milestone.