Rhapsody Code Generation with dependencies takes longer on single processor
Attempts to run code generation with dependencies against the same model from same environment using IBM Rationa Rhapsody version 8.0 or above might result in taking much longer time than previous version.
This phenomenon is related to a new feature, Parallel Code Generation, which was introduced on Rhapsody version 8.0.
By default, property CG::General::ParallelCodeGeneration is set as BasedOnNumberOfCores.
If your model is using default settings and your environment has multiple processors, the tool would launch multiple rhapsodyCL.exe processes to generate code parallelly.
Parallel Code Generation distributes the whole work into several processes, each re-loading the relevant part of the model. Because of this behaviour, enabling parallel code generation feature would probably worsen the tool performance if your environment only has one processor.
Rhapsody version 8.0 or higher
Resolving the problem
To avoid the performance issue, you need to disable Parallel Code Generation feature by setting the value of CG::General::ParallelCodeGeneration to Disabled.
Alternatively you can try another setting UserDefinedNumberOfProcesses combining with property CG::Geneal::UserDefinedParallelProcesses to intentionally request the tool launching multiple rhapsodyCL.exe processes regardless only one physical processor is available.
More support for:
Software version: 8.0, 8.0.1, 8.0.2, 8.0.3, 8.0.4, 8.0.5, 8.0.6, 8.1
Operating system(s): Windows
Reference #: 1677491
Modified date: 16 July 2014