Towards a useful quantum repeater

Building the first large-scale quantum network is a highly challenging endeavor. Besides the choice of the most promising hardware platform being contentious, even if there were a clear winner, the precise requirements for its realization would be unknown. Here, we conduct a detailed study on connecting the dutch cities of Delft and Eindhoven via SURFNet's fiber network using a processing node repeater. We investigate setups based on NV centers in diamond, trapped ions and a very general abstract model, taking into account real world fiber data and possible physical locations for hardware placement. To perform this study we have developed a general methodology using NetSquid, a discrete event simulator for quantum networks, as well as optimization tools based on genetic algorithms. Specifically, we show the minimal improvement necessary in relation to present-day hardware parameters to connect the two cities. We compare this real world architecture study with the idealized setting typically considered in the quantum repeater literature, and quantify the difference in requirements. Our analysis provides a Blueprint for realizing a quantum repeater in practice, and showcases limitations in existing studies of idealized situations. We make our simulation and optimization tools freely available, in the hope that this can contribute to bringing the quantum internet online as fast as possible.