beanstalkd might be what you are looking for. It's ancient but solid af. I know of systems still running today that use it. For something more modern, check out Nats—and if you are using Go, you can embed the server in your executable.
I really like NATS, but it's worth noting that it's only really "the SQLite of queues" in the context of their Go client, which can embed a NATS server into an application.
Otherwise I would really say it is more like the Postgres of queues (still great, but potentially not what OP is looking for)
I’m sure there are better solutions on paper and for performance under common workloads. But if SQLite is good enough for your use case, it’s good enough for your use case.
https://github.com/yudhasubki/blockqueue
Literally, the sqlite of queues.
SQLite is the SQLite of queues. Simply use a table per queue. And use your choice of SQL to implement the types of queues you want.
beanstalkd might be what you are looking for. It's ancient but solid af. I know of systems still running today that use it. For something more modern, check out Nats—and if you are using Go, you can embed the server in your executable.
If you need them for IPC within same host then POSIX queues (see https://pubs.opengroup.org/onlinepubs/7908799/xsh/mqueue.h.h...) are a thing.
Redis can be used as simple message queue as well.
Try NATS @ https://nats.io
It's a single binary and can operate as a message bus, persistent queue, KV store, object store, provide services, and so on.
I really like NATS, but it's worth noting that it's only really "the SQLite of queues" in the context of their Go client, which can embed a NATS server into an application.
Otherwise I would really say it is more like the Postgres of queues (still great, but potentially not what OP is looking for)
ElasticMQ: https://github.com/softwaremill/elasticmq
RabbitMQ
ZeroMQ - https://zeromq.org/
Notably not a persistent queue, correct?
this is redis for me.
/var/queue/
Serious engineering question:
Will SQLite work for your application?
I’m sure there are better solutions on paper and for performance under common workloads. But if SQLite is good enough for your use case, it’s good enough for your use case.
It probably isnt. Good luck.
Was wondering this for OP as well, I've used a database as a queue for small projects in the past and it was fine.
[dead]
[dead]