Hi,
I'm experimenting with RabbitMQ (AMQP) by using it to handle sequential chains of background tasks. Tasks include comparing data from datastores, updating items, making external API calls, etc. Units of work may follow different paths through a branching structure I am constructing with RabbitMQ.
Meanwhile, most AMQP examples I've seen, including the RabbitMQ tutorials, describe tasks at one level of depth.
The difference between official examples and my experimentation concerns me that I am misusing AMQP. So, my question:
Is it preferable to combine all operations in independent tasks, keeping RabbitMQ utilization one-level deep, or is it just as well that tasks be chained in RabbitMQ?
Thank you for any comment,
Gallagher