Managed Services Engine

During the last couple of weeks I've been experimenting applying Microsoft's Managed Services Engine (MSE) to SOA Governance scenarios that I've typically addressed with HP-Systinet2, SoftwareAG and SOASoftware. Although still in very early stages (the first CTP was just announced couple of weeks ago) we can already appreciate some of the core ideas behind this technology.  So why am I so excited about this effort?

The first reason is that I am a strong believer on the possibilities of a SOA Governance technology based on WCF.  If you follow this blog, you know that during the last year I've been doing a lot of work with the combining some of the lead SOA Governance frameworks with WCF and BizTalk. Multiple times I ran into scenarios in which the SOA Governance technologies were insufficient to manage some of the key features exposed by the "governed" services; specifically in the case WCF applications. The fact of the matter is that a SOA Governance technology is only as good as the Web Services engine behind it. Let's use a couple of  examples , in order to govern interoperable security policies a SOA Governance technology should support the latest generation WS-* protocols which includes WS-Policy, WS-Security, WS-SecureConversation, WS-Trust, etc.  Almost all the SOA Governance vendors in the market, arguably HP-Systinet could be seen as an exception, face tremendous challenges on this aspect mostly because they don't include a strong Web Services engine behind the governance technology. A similar case applies when comes to modeling Web Services components such as contracts, bindings, policies, etc which are better expressed as part of the programming model of a Web Services technology than as part of the SOA Governance framework itself. In the case of MSE, the fact that its completely based on WCF notably increases the options for addressing key SOA Governance scenarios such as WS-* protocols governance, contract management, policy management, binding management, etc.

The second reason is obvious: Microsoft's Connected Systems stack needs a SOA Governance technology. Although System Center provides some interesting options for implementing runtime governance there are lots of scenarios that needs a more domain specific (I love that term J) technology that leverages the WCF programming model.  With more and more Microsoft products adopting WCF as its messaging engine; the need of establish a governance strategy that can fully leverage the rich set of capabilities of the WCF programming model it's becoming a fundamental requirement of Enterprise applications.  

I am not sure what is going to be the future of MSE in the Connected Systems stack; but what is undeniable is that the ideas behind this technology should serve as the core for some of the most important components of Microsoft's SOA Governance story.

In the next days I plan to blog more details about the scenarios on which I've been playing with MSE.