Burlesque – A message queue server in Go with an HTTP API
1–5 of 5 posts
Re: Burlesque – A message queue server in Go with an HTTP API
#2Can someone explain the benefit of this over using a list in redis and using a blocking pop. Only thing I can think of is that it uses HTTP so you may be able to get around some port restrictions in your infrastructure, but that is probably a very uncommon case.
Re: Burlesque – A message queue server in Go with an HTTP API
#3Can someone explain the benefit of this over using a list in redis and using a blocking pop. Only thing I can think of is that it uses HTTP so you may be able to get around some port restrictions in your infrastructure, but that is probably a very uncommon case.
I don't know anything about this project, but Redis is not a message queueing system. It kind of baffles me why people insist on using it as such when there are plenty of systems out there designed with real pub/sub delivery semantics. In the Go ecosystem I really like NSQ (http://nsq.io).
Re: Burlesque – A message queue server in Go with an HTTP API
#4Can someone explain the benefit of this over using a list in redis and using a blocking pop. Only thing I can think of is that it uses HTTP so you may be able to get around some port restrictions in your infrastructure, but that is probably a very uncommon case.
Redis can behave as a message queue, but it is somehow more complex than a blocking pop. antirez wrote a nice article[1] about it, and I believe RestMQ[2] uses this approach.
Re: Burlesque – A message queue server in Go with an HTTP API
#5Can someone explain the benefit of this over using a list in redis and using a blocking pop. Only thing I can think of is that it uses HTTP so you may be able to get around some port restrictions in your infrastructure, but that is probably a very uncommon case.
This server was built for a specific purpose: storing huge amounts of critical delayed job descriptions.
Publishing and retrieving jobs is done via HTTP as it's the easiest transport protocol for most stacks.
KyotoCabinet database is chosen for its capabilities to tune the speed and the size of the storage depending on the nature of its data and the flow of the app.