Neuron ESB User Network

The Service Bus for the Connected Business

CM

Services Aggregation with Neuron ESB today and tomorrow

Many users are familiar with utilizing Neuron for Location Transparency and Services Mediation as well as standard EAI. And, a fair number of users are familiar with using Neuron's pipeline features such as WF or Notifications or Transformation.

Very few users however are aware that you can achieve patterns with Neuron that seem to require a classic orchestration engine. I was asked recently about one of those patterns Scatter-Gather and provided a solution that was targeted to that particular use case.

I realized that this solution was not immediately intuitive because Neuron is a more obvious Messaging Engine today than it is an EIP engine and decided that I would build a more generalized example so that more folks would understand how to do such things.

I've uploaded the code so that you can work with it yourselves but the basic pattern works like this. The Party in the Client Connector contains a WF which splits the incoming request and uses a proxy for the original Party to run pipelines which use CBR to determine message paths and action and then publishes the requests to the bus with Semantic.Request.

Did you catch that? When you read the code and the ESB you'll see it's simpler than that Faulkneresque sentence.

The replies are then gathered and returned to the bus with a Semantic.Reply.

So, essentially it's a Party in a Client Connector using a WF to talk to the bus and itself!

That's today.

In the land of tomorrow... which is coming very soon some EIPs will be baked into our new services engine and you will be able to compose solutions graphically. In addition, our fearless leader Marty Wasznicky has targeted persisted WF for the 3rd qtr. So many classic orchestrations and long running process scenarios will be available within the year.

We're not losing our roots though. Neuron was built with messaging in mind and the mad scientist Mr. Jesse Ezell has been hitting some numbers that will make fans of speed stand up...

Anyways, before I get too excited back to today....Please run the scenario by first unpacking the Zip and loading the aggregotor.esb in offline mode. Make sure you fix up the path to the WF in the integration tab if it differs. Fix the references to link to your ESB, compile and hit CTRL-F5 to start the example services. Finally switch your active .esb to point to aggregator.esb and you should be ready to go to test view in VS and choose RunExample.

This is a somewhat advanced scenario so feel free to ask questions...There's actually code in there to do aggregation in the WF with our Client Connectors instead of the .NET API and you can also add multicasting with one line but I’ll leave those as the dreaded exercise for the reader!


The code is here: AggregationViaWFExample.zip

Views: 0

Comment

You need to be a member of Neuron ESB User Network to add comments!

Join Neuron ESB User Network

Neuron ESB Product Support Forums and Communities

Latest Activity

Profile Icon

Sending messages to a Client Connector from an C# Console App

I have an issue with sending messages to a Client Connector from a C# test app. The main issue is that  i never receive a response from Neuron and the Connection Times out. I'm using the sample code from the Scatter-Gather Pattern sample included in the documentation If anybody has any hints on what i'm missing, it would be appreciated.static class Utility {const string _propName = "NeuronServiceList";const string _ns = "urn:xmlns:neuronesb-com:soapheaders";const string _msg = @"…See More
Discussion posted by John Ryan Friday
Profile Icon
ThumbnailThumbnail
AMS and John Ryan joined Neuron ESB User Network Friday
Profile Icon

Scalability MSMQ & Deployment Groups

I have been trying to work out how to configure MSMQ for high availaibility. I think that I need to use Zones, & Deployment Groups to configure the server deployment.  But I am unsure how  I would go about this and what the perfromance impacts are. So let me give you the scenario that I am looking at and you can then point me in teh direction of resources that I need to be able to resolve it. I have 3 High throughput topics that need reliable messaging Service1Service2Service3 Each of these…See More
Discussion posted by Alixx Skevington Feb 7
Profile Icon
Alixx Skevington is now a member of Neuron ESB User Network Feb 6
Profile Icon
Umesh Kumar Maurya is now a member of Neuron ESB User Network Jan 24
Profile Icon
David Gnabasik is now a member of Neuron ESB User Network Dec 16, 2011
Profile Icon

Explanation of Pipelines

Hi Guys,               I need more documentation/info on how pipelines work. It appears that a context is the message that is passing though the pipeline. I want to use a code step to generate a new message to that will be passed to a service step. But I can not find any detail on how to use the context object. Regards,                   AlistairSee More
Discussion posted by Alistair Rigney Dec 7, 2011
Profile Icon
Alistair Rigney is now a member of Neuron ESB User Network Nov 25, 2011

Badge

Loading…

© 2012   Created by Neuron Admin.   Powered by .

Badges  |  Report an Issue  |  Terms of Service