Describes the architecture of the procession
server.
- A
procession
server manages jobs using a backingredis
database. - One or more
client
instances push jobs to the server to be processed. - One or more
client
instances consume jobs from the server.
As the server gets implemented, this'll be fleshed out more. At a high level, here are the architectural goals:
- Server
- Support connecting to a pool of replicated redis databases.
- Support running multiple
procession
servers against the same pool.
- Clients
- Any client can be a job producer or consumer.