In this paper we show how to extented the computing power available to Scilab in term of targeted resources: we present how to interface Scilab with an existing middleware to take benefit of its resource management system. Scilab is a useful environment in the context of high power computing, allowing an easy development of applications. However, when application data grows, the lack of parallel constructs and the complexity to manage heterogeneous resources limit Scilab’s usage for real cases. Hence, our goal is to distribute Scilab’s tasks over a distributed computing platform (e.g., Grid or Cloud platforms) to leverage the provided computing power. To do so, we chose the Distributed Interactive Engineering Toolbox (DIET) middleware. DIET is designed for high-performance computing in heterogeneous and distributed environments, which can easily be used for parallel computations. In this paper we explain the process to integrate Scilab with an existing middleware. We hope that this article will be useful for users of this solution, or for developers who want to integrate another middleware. We first describe the DIET middleware and the Scilab’s functions we used in this work. Then, we describe the proposed architecture: from the layer on the user’s side, down to the execution stack on the server side. Finally, we provide examples on a real computation, and as a proof of concept we show how to use this solution.