Technical Conference Day 2

As the day is passing by more details are being revealed on the Technical Conference break-out sessions. Here is a summary of the keynote and some topics for Solution architects and Developers:

  • Various models will be available for deploying AX. Current version is the Ready-To-Web release, which is in line with Microsoft’s Cloud First strategy for brand new implementations. Later on Azure Private Cloud deployments will be the next step down the ladder, after which the On-premise deployments will follow. This needs to be synchronized with other technologies to be available, namely Windows Server 2016, SQL Server 2016 and Azure Stack on-prem
  • The main goals with the Cloud First strategy are to offload responsibilities from Partners for infrastructure sizing and maintenance, application platform patching and seamless deployment of new functionality, and lastly sealing down modules to make it more configuration-based rather than doing bespoke changes to go for an SaaS structure
  • Form styles are now referred as Patterns and they became mandatory. The number of pattern templates have been greatly increased, and they fall into the categories of Form Pattern (top level) and Subpattern (lower levels). Special cases are supported as Custom Patterns, however the performance, responsiveness and upgradeability are up to the implementor
  • A developer box now requires minimum 16 GB of RAM due to the highly optimized compiler (xppc), which can do a full compilation including cross-reference updates in 9 minutes on a VM which meets the system requirements
  • The X++ Language Toolkit (XLNT) is completely rewritten in C#, and produces native MSIL managed code that executes server-side. The source code is now stored in a file structure using XML format
  • Ignorelist could be used to partly skip code with problems during the migration, to be able to generate the assembly for the pieces that can compile and are ready for testing
  • Packages/models are now making the traditional layering obsolete, we could think of it like there are unlimited number of layers
  • The overlayed code only stores a differential delta compared to the model below, and layering is now much granular (i.e. Form datasources are no longer a single object that we cannot even compare)
  • Code changed on the Application suite or any other models flag them for code conflict, which we could merge in Visual Studio very intuitively
  • xRef is now sitting in it’s own database
  • Different Best Practice rules could be applied per model, and they could also be extended with our own rules
  • A lot of interesting features came in from the .Net framework, just to name a few: variable declaration without type using VAR keyword, CONST and READONLY keywords for static variables, access restrictions on variables using PUBLIC/PRIVATE/PROTECTED, THIS.method/variable calls for clarifying scope, TRY-CATCH-FINALLY, USING statements and clauses, _Extension methods, code-based and declarative event handlers

Many points above are already available on the AX wiki, here is an overview for the development environment feature changes:

https://ax.help.dynamics.com/en/wiki/using-new-x-and-debugger-features-in-ax7/

DAXRunBase / 2016-02-25 / AX 7
Tags: ,

Comments

  1. Technical Conference Day 2 - Microsoft Dynamics AX Community

Leave a Reply

Your email address will not be published / Required fields are marked *