Scientific applications are still getting more complex, e.g. to improve their accuracy by taking into account more phenomena. Moreover, computing infrastructures are continuing their fast evolution. Therefore, software engineering is becoming a major issue to achieve easiness of development, portability, simple maintenance, while achieving high performance. Software component model is a promising approach, which enables to manipulate the software architecture of an application. However, existing models do not capture enough resource specificities. This paper proposes a low level component model (L2C) that supports directly native connectors such as MPI, shared memory and method invocation. L2C is intended to be used as a back end by a ''compiler'' (such as HLCM) to generate an application assembly specific to a given machine. This paper shows on a typical domain decomposition use case that \llc can achieve the same performance as native implementations, while gaining benefits such as enabling resource specialization capabilities.
