April 2009 - BizTalkGurus

Screencast: Windows Workflow – Creating Custom Context Bindings

by community-syndication | Apr 30, 2009 | BizTalk Community Blogs via Syndication

I’ve recently published a free screencast on the basic ideas around context management with workflow services.

wf-custom-context-binding

Microsoft ships several bindings that support context management, but they cannot possibly cover all of the scenarios you might need. In this screencast, I'll show you how to build your own custom context binding using a netNamedPipesContextBinding as an example.

Check out our growing collection of free .NET screencasts and videos. Subscribe to the Pluralsight feed to be notified when new screencasts are published. Also, check out our growing library of online .NET training courses — see what you can learn with Pluralsight On-Demand!

Screencast: Windows Workflow Foundation workflow service context basics

by community-syndication | Apr 30, 2009 | BizTalk Community Blogs via Syndication

I’ve recently published a free screencast on the basic ideas around context management with workflow services.

wf-service-context-basics

This screencast walks through the basic ideas of managing context in workflow services and introduces several concepts that will be used for deeper exploration in other screencasts in the series.

Designing Services for Management & Scale with Dublin Video Online

by stephen-w-thomas | Apr 30, 2009 | Stephen's BizTalk and Integration Blog

A few months ago Kent Brown and I presented on Dublin (the new enhancements to Windows for Workflow Hosting, Monitoring, and Tracking) at the 2009 SOA and Business Process Conference in Seattle Washington.

This session was recorded and is now online. You can view this video here: http://soaconference2009.spaces.live.com/blog/cns!FA6FC7F5DB1C07!135.entry

The abstract for the session is below.

Abstract: Dublin is the code name for the new Windows Application Server features inside Windows. In this session we will take a look at how to build a service to leverage the management tools inside Dublin to gain insight into the health of running services. We will show how simple it is to configure Dublin to custom track custom values inside message payloads. Once tested, deployment and migration of the service to a new environment will be demonstrated.

Watch for more to come from me on Dublin in the up coming months.

BizTalk 24*7 Publishes blogs in a structured Index

by community-syndication | Apr 30, 2009 | BizTalk Community Blogs via Syndication

Down at BizTalk 24*7 Saravana Kumar has been working hard to organise a whole collection
of BizTalk articles from us in Cyberspace

He’s done a great job!

Check it out – http://blogdoc.biztalk247.com/

and yours truly is here – Mick
Badran

BizTalk RFID Workshop Review

by community-syndication | Apr 30, 2009 | BizTalk Community Blogs via Syndication

Earlier this week I attended the RFID workshop held at Microsoft’s Reading Offices. I thought Id write up some notes and thoughts I’d made about the event.

Content/Trainer

The course was based on the course developed by the guys at Breeze (http://breeze.net/default.aspx). The material itself is excellently written and is easy to follow. It also has some nice little bits of humour which mean it is not the common/boring material you get on some courses. The material itself lets you build and enhance an RFID solution through out the course of the labs and there are about 10 labs and the course was very hands on.

The trainer for our course was Jeff Johnson from Microsoft in the UK and he delivered the course very well and was able to add lots of additional information based on experiences from work he had done.

The course itself was a 2 day workshop and was free.

Freebies

There were some excellent freebies on this course:

A copy of the BizTalk 2009 RFID book
A USB RFID reader/writer
A small bag of RFID tags
A BizTalk VPC (2006 R2) with everything on to do additional study away from the course

Attendees

The attendees for the workshop gave us an excellent mix of BizTalk people who had done little or no RFID work and RFID specialists who hadn’t done much with BizTalk.

This resulted in lots of discussion and interesting points based on peoples differing experiences. It was also good to see some people from the UK SBUG user group there.

My Random Thoughts based on what I learnt on the course

I think the first thing to say is that I have wanted to look into the RFID aspects of BizTalk for quite a while, but the combination of me not having any RFID work in the pipeline and my expectation that it would be quite a steep learning curve and that I didn’t have the hardware I would need to play around with it had put me off doing this for a long time. It just kept getting reprioritised.

Jeff had actually contacted me to promote the workshop to our user group members and I thought it was a good time to do some training and being free it ment I would only lose the cost of the days off work.

To my surprise RFID work with BizTalk is sooooo much easier than I expected, so much so that I had to question if it was appropriate to call it BizTalk being so easy.

I don’t really want to go into too much discussion about the aspects of RFID but some things which I feel will catch people’s interest to look more into this might be:

Traditionally RFID work had the typical complications of working with hardware vendors. You needed different API’s for each vendors readers etc. BizTalk RFID abstracts this and gives you a standardised API which allows you to communicate with any supported vendor. You could really see the benefits of this and you can code directly against this API from .net applications if required
BizTalk RFID is actually a separate module to the main BizTalk installation. The course really discussed setting up RFID on “edge” servers which would be configured to work with a set of readers. You could then send your events to a central hub BizTalk installation to interact with LOB applications if required
Following on from the above comment it really looks like you would use RFID with the branch edition of BizTalk on your edge server. This significantly reduces the cost of setting up solutions and also if I remember correctly you don’t need a hub BizTalk enterprise edition server if you don’t need one. This allows customers who may use other vendors for their integration technologies to still take advantage of BizTalk RFID as a low cost solution to work with their current products, it also offers SME companies a cheap option and as they grow it gives them the option to bring “proper” BizTalk into the picture later.
BizTalk RFID doesn’t require the traditional BizTalk skill set. There is some configuration requirements to configure your devices and RFID processes, but when it comes to custom processes the components are written in .net code in a similar way to BizTalk pipelines and pipeline components (but much simpler!!)
The RFID module gives you an execution environment for processing your RFID events and you can have multiple processes which will pick up the same events depending on how you configure your bindings from devices to processes.
The configuration/administration side of things is quite BizTalk like but with out any of the SQL Server requirements
There are some testing things that come with RFID which allow you to test your processes without having to have an RFID reader

I’m sure that there is a whole lot of other stuff that I should say but just doesn’t come to mind at the moment.

In terms of recommending the course I would say that it is a very good course and it has really excited me about the possibilities of doing BizTalk RFID solutions.

I have invested 2 days of time and I’m coming away with a good understanding of this and I think with very little effort I can setup a pretty powerful demo to show my customers.

I believe Microsoft might be running this workshop again in the Autumn or if you are outside the UK I’m sure Breeze can advise you how to do this course.

WebSphere loves Windows, enough said

by community-syndication | Apr 30, 2009 | BizTalk Community Blogs via Syndication

Recently, Greg Leake (Senior Director Developer Platform at Microsoft) performed a series of IBM written benchmarks. The results of the benchmarks demonstrate a significant cost savings of running WebSphere on Windows on HP Blades as opposed to running the same workload on IBM's Power 570 hardware. This savings is not only significant but the overall performance in terms of transactional throughput was significantly better than on the Power system. In addition, the tests show the ability to scale the overall system in a granular fashion. The net findings of these tests is that an organization has the possibility to run their WebSphere application workloads on Windows on Intel based hardware for markedly lower costs with improved performance as compared to Power systems. You can find an overview of the work at http://www.websphereloveswindows.com/.

For more in-depth information on the studies, check out http://www.microsoft.com/windowsserver/mainframe/whoknew/, as well as the information found on MSDN around the .NET StockTrader Sample Application. And you can find Greg Leake’s blog here.

Webcast – BAML – A DSL for BAM

by community-syndication | Apr 30, 2009 | BizTalk Community Blogs via Syndication

I never really “got” the idea of the need to build a textual DSL when I was first introduced to MGrammar. The light really switched on when I looked into developing a DSL that would make it easier for developers to create BAM activities.

The BAML language only took a couple of hours to develop. I had experimented with simple text based DSLs before, so this was my first “real” language.

I have recorded a 20 minute webcast showing how the language works, and how it can be used. If you want to experiment with it yourself, the language is here.

module BloggersGuides

{

language BAML

{

syntax Main = Activity;

syntax Activity =

ActivityToken

n:NameToken

‘{‘

p:List(PKI)

‘}’

=> { activity { n,{ p }} };

syntax PKI = Milestone | Integer | Decimal | Text;

syntax Milestone = t:MilestoneToken n:NameToken ‘;’

=> { t, { n } };

syntax Integer = t:IntegerTokenn:NameToken ‘;’

=> { t, { n } };

syntax Decimal = t:DecimalTokenn:NameToken ‘;’

=> { t, { n } };

syntax Text = t:TextTokenn:NameToken ‘;’

=> { t, { n } };

syntax List (Element) =

e:Element => { e } |

list:List(Element) e:Element => { valuesof(list), e };

token NameToken = (‘A’..’Z’ | ‘a’..’z’)+;