Data synchronization using akka-camel

[Gonzalo Vasquez]

Hi folks!

We are currently planning a solution to a problem, where I think that akka, camel or a mix of both can be handy, and I'd like your recommendations about it:

Environment:

- AIX HA Cluster with 2 cluster nodes nowadays, solution must be smart enough to be platform independent as we have other scenarios with Linux & MS Windows also.

- IBM WebSphere Application Server v6.1 on each node, JBoss, Glassfish and Weblogic on other scenarios

- IBM Content Manager OnDemand (CMOD) v8.5 on each node (document repository)

What we need is to achieve consistency between the cluster nodes in terms of repository content, files can be either delivered to either nodes independently, or to a single on, it's our choice, but the big issue here is that once loaded onto one machine, it must also be loaded on the other.

Typical problems are:

- One machine is down so data gets loaded only on a single node, how does the other ensures catching up?

- Data arrives to a single machine, same scenario above

- Loading fails on one of the nodes

FYI: data gets loaded (or unloaded in case of rollback needed) via command line execution, as there's no API available for such processing

We've already implemented an application to try to avoid those problems, but it's total chaos!

Any ideas on how to accomplish such consistency / synchronism?

Gonzalo Vásquez Sáez

Gerente Investigación y Desarrollo (R&D)
Altiuz Soluciones Tecnológicas de Negocios Ltda.
Av. Nueva Tajamar 555 Of. 802, Las Condes
(56-2) 335 2461
gvasquez@altiuz.cl

http://www.altiuz.cl

 

[√iktor Ҡlang]

On Wed, Oct 24, 2012 at 5:07 PM, Gonzalo Vasquez <gvas...@altiuz.cl> wrote:

Hi folks!

We are currently planning a solution to a problem, where I think that akka, camel or a mix of both can be handy, and I'd like your recommendations about it:

Environment:

- AIX HA Cluster with 2 cluster nodes nowadays, solution must be smart enough to be platform independent as we have other scenarios with Linux & MS Windows also.

- IBM WebSphere Application Server v6.1 on each node, JBoss, Glassfish and Weblogic on other scenarios

- IBM Content Manager OnDemand (CMOD) v8.5 on each node (document repository)

What we need is to achieve consistency between the cluster nodes in terms of repository content, files can be either delivered to either nodes independently, or to a single on, it's our choice, but the big issue here is that once loaded onto one machine, it must also be loaded on the other.

Typical problems are:

- One machine is down so data gets loaded only on a single node, how does the other ensures catching up?

- Data arrives to a single machine, same scenario above

- Loading fails on one of the nodes

FYI: data gets loaded (or unloaded in case of rollback needed) via command line execution, as there's no API available for such processing

We've already implemented an application to try to avoid those problems, but it's total chaos!

Any ideas on how to accomplish such consistency / synchronism?

This is completely the opposite of what Akka is trying to achieve: eventual consistency and asynchronous.

You should probably look into so kind of distributed storage.

Cheers,

√

 

Gonzalo Vásquez Sáez

Gerente Investigación y Desarrollo (R&D)
Altiuz Soluciones Tecnológicas de Negocios Ltda.
Av. Nueva Tajamar 555 Of. 802, Las Condes
(56-2) 335 2461
gvasquez@altiuz.cl

http://www.altiuz.cl

 

--
>>>>>>>>>> Read the docs: http://akka.io/docs/
>>>>>>>>>> Check the FAQ: http://akka.io/faq/
>>>>>>>>>> Search the archives: https://groups.google.com/group/akka-user
---
You received this message because you are subscribed to the Google Groups "Akka User List" group.
To post to this group, send email to akka...@googlegroups.com.
To unsubscribe from this group, send email to akka-user+...@googlegroups.com.
Visit this group at http://groups.google.com/group/akka-user?hl=en.
 
 

--
Viktor Klang

Akka Tech Lead

Typesafe - The software stack for applications that scale

Twitter: @viktorklang