A wide variety of new applications is being realized to exploit the immense computing capabilities of heterogeneous many core platforms, concurrently running multiple parallel tasks on the available cores. This is opening up to a series of technological challenges in the real-time and embedded computing market, ranging from the parallelization of existing applications, to the simultaneous elaboration of multiple sensor data, to the need for predictable timing guarantees of applications requiring a prompt interaction with the user/environment. The problem is that it is difficult to provide predictable timing guarantees when multiple cores may contend for shared resources, like memory, buses and I/O devices. A large number of results have been proposed by the real-time research community to address these issues: multi-core scheduling algorithms, schedulability tests, predictable execution models, etc. However, we are not aware of any industrial-level application that brings this wide algorithmic background to an industrial setting. We show how these problems impact real scenarios and how they were solved during the HERCULES project which came up with an integrated framework in order to achieve a predictable performance on top of low- power multi-core heterogeneous platforms. Moreover we illustrate how the framework is divided and all the techniques used in order to have such a real time system.