Even camel-core which the full EIP implementation is just about 2M jar, you can use it as library, it doesn’t mean you have to embed it in you application. It really dependents on your use case: If you want to change and deploy the camel route repeatedly, you may not want to use camel in a embed way. If you are using some other camel components which are not included in the camel-core, you may consider if it could be more effective to run a separated JVM to host the camel route. -- Willem Jiang
Red Hat, Inc. Web: http://www.redhat.com Blog: http://willemjiang.blogspot.com (English) http://jnn.iteye.com (Chinese) Twitter: willemjiang Weibo: 姜宁willem On June 13, 2014 at 7:05:33 AM, blay ([email protected]) wrote: > We're a defining an architecture for our company applications, so that > architecture has to take care of integrations with other > applications/systems. The first thing that came our mind was to centralize > all integrations in an ESB (Mule), which would run in standalone mode. But > new requirements appeared and the thing is that every application has to > manage its own integrations (which are not defined yet, but they shouldn't > be very complex). Now we´re evaluating using Apache Camel or Mule but in an > embebbed scenario. Discusing about it with some mates, it isnt very clear to > us which is the best (or more suitable) way to set up this architecture. At > my opinion, as its responsability of each application, I would integrate > Camel directly in my application (as a library); but they say it´s a better > option to deploy Camel in a separate project. These are the scenarios, as I > see them: > > My application with Camel embebbed. If my application has to invoke a > webservice, for example, I just code it and send it (From(...).to(...) etc) > > My application and another application with Camel embebbed. If my > application has to invoke a webservice, and I want to manage all > integrations through Camel, I think I have to call the camel project (by > JMS, or calling an interface it exposes), define a route in that project > which says: when I read from X (the interface im calling) call this WS. I > mean, it adds more complexity when I think it´s not needed. > > Probably I missunderstanding how Camel really works so I´ll be glad to hear > what I´m wrong about ;) > > > > -- > View this message in context: > http://camel.465427.n5.nabble.com/Best-suitable-architecture-using-Camel-tp5752219.html > > Sent from the Camel - Users mailing list archive at Nabble.com. >
