This work deals with the problem of developing embedded multi-core systems, under time-to-prototype and high performance constraints, by exploiting reconfigurable logic. In particular, the paper focuses on the early analysis activities, performed by means of native simulation technologies, and then on the full development of an embedded multi-core platform composed of four LEON3 soft-processors and able to support the execution of OpenMP based applications. Moreover, in order to evaluate software execution time at run-time, a distributed hardware profiling mechanism has been inserted into the final implementation to monitor the whole system. The final goal is the definition and the exploitation of a high-level design and validation methodology able to exploit both a well-known parallel execution paradigm and a run-time system monitoring service. Correctness and performance of such a platform have been evaluated by means of specific benchmarks.