Performance modeling of partially reconfigurable computing systems

Reconfigurable systems have become very popular because of their impact in increasing performance improvement for executing a number of applications such as morphology, image compression, etc. These systems provide a general platform for executing applications. However, effectively using the full potential of these systems cars be challenging without the knowledge of the system´s performance characteristics. A general analytical model is developed in this paper for reconfigurable computing systems based on queuing theory. The reconfigurable system is composed of a host processor coupled with a reconfigurable hardware such as an FPGA.