Blog

Posté le dans Blog
After Java Productivity tour 3 weeks ago, I could participate to the first edition of "Jeudis des modèles". Theses meeting take place in EMN Nantes every two months and deal with the Model Driven Engineering.So, thuesday there was two presenters for this first time :Nicolas Rouquette from the JPL, NASAAnd Patrick Albert from ILog (IBM)The first talk (from Nicolas) introduced the use of models in the JPL and especially why NASA came to use the models in its software design process. Nicolas, who defines the Software design methodology, explained the evolution in the way of create software that drive the lander sent on mars. Of course, this kind of software must be very robust and it design became more and more harder this last years.So JPL had to improve again and again its design process and with the growing complexity of this process, they came to using models since 10 years now.The talk of Patrick describes the evolution of software design in ILog. During the ILog evolution, they real...

Posté le dans Blog
One of the main features of the standalone Acceleo engine is to provide multi-thread support. This allows us to divide the code generation time according to the count of available processors, effectively cutting the generation in half on some machines. To share the work between threads, the input model is divided into groups of objects; each group is then queued till a thread can process it.There are two parameters to tweak the multi-threaded generation. The first one is the number of threads to run simultaneously, the default value being the number of available processors plus one. This allows us to use all available processors, the one supplementary thread used to cover synchronization waiting time. While some threads are waiting for another one, one processor is free to run the thread which was waiting for an idle processor.The second parameter is the number of objects per group, the default being 20 objects per group. This parameter is very important since its allows you to find th...

Posté le dans Blog
Thuesday took place in Nantes (La Beaujoire) the JavaEE Productivity tour. It was a presentation of three companies Obeo, ProxiAD and SpringSource dealing with their feedback on productivity in the JavaEE world.First, Arnaud Buisine, from ProxiAD, presented the current state of his company about project industrialisation. His presentation deals with many topic like :The choice of frameworksThe choice of code generationThe quality processThe developper post : settings and building...Then comes Julien Dubois from SpringSource. His part dealt with the productivity in the framework point of view. He talk about :Spring what it does and what it will do !GroovyGrailsFinally, Etienne from Obeo, shows our tooling dedicated to JavaEE productivity. It was the moment to talk about :Acceleo and Acceleo ProObeo designerAnd Obeo NetworkAfter this 3h30 of presentation, It was the time to lunch and discuss about this topic. Interesting topics make interesting audience and interesting audience implies e...

Posté le dans Blog
Here we go !I've just open my blog dedicated to my work in the Eclipse modeling world.Currently, I work on the EEF project which aim at providing a new framework to the Eclipse user to design advanced editing components. Once these are designed they can be include in SWT/JFace elements as tabbed properties sheets or JFace wizard.Here is two examples of the EEF generation result :In a tabbed properties sheet page :In a JFace wizard :After a good Proposal Phase (with some propositions of several interested parties), we currently work on the Creation Review for EEF. Thanks to everybody who help us to create this exciting project !Lien d'origineAuteur d'origine: Goulwen...

Posté le dans Blog
There is a new feature in Acceleo 2.5. This feature allows users to profile the generation process of Acceleo. To activate the profiling you should check the profiling checkbox in the launch configuration of the chain.The result of the profiling session is saved as a model, it's very useful for performance comparison (using EMF Compare) or report creation in batch environment (using Acceleo :) ). Basicaly the model is the execution tree of Acceleo, each node being a step of the generation process. Those nodes give statistics like the time spent in the node and its subtree, the percentage of time it represents, and the number of times the element have been run by Acceleo. You can also find the list of EObjects for each node in the outline. The model can be sorted by chronogical order or by the time spent as shown on the screenshot. It can help in the understanding of the generation process and finding the hotspot(s). Once this hotspot is found you can open the template editor on thi...