Choose broker and integrate with server

[kylejb]

What

We need to poll NDBC endpoints hourly for latest buoy weather readings and persist this information in our databases.

Tasks

1. Choose an appropriate broker (e.g., RabbitMQ).
2. Create "hourly" task to update forecast data.
   • should this apply to all buoys or a subset of buoys?

Why

Celery, along with an appropriate broker (e.g., RabbitMQ), would allow us to scale without impacting application performance. Specifically, this should allow us to fetch and persist large amounts of data as a separate process, which would avoid negatively impacting end-user experience.

[kylejb]

FYI - this is not currently a priority; it is necessary before deploying to production.