This work focuses on testing the consistency of distributed and adaptable systems. In this context, Runtime Testing which is carried out on the final execution environment is emerging as a new solution for quality assurance and validation of these systems. This activity can be costly and resource consuming especially when execution environment is shared between the software system and the test system. To overcome this challenging problem, we propose a new approach to design a resource aware test architecture. We consider the best usage of available resources (such as CPU load, memory, battery level, etc.) in the execution nodes while assigning the test components to them. Hence, this work describes basically a method for test component placement in the execution environment based on an existing model called Multiple Multidimensional Knapsack Problem. A tool based on the constraint programming Choco library has been also implemented.