Content-based publish/subscribe systems provide a useful alternative to traditional address-based communication due to their ability to decouple communication between participants. Removing the dependency between the interacting peers makes it well adapted to large scale distributed applications such as news delivery, stock quoting, traffic control, on-line games, and dissemination of multimedia contents. It has remained a challenge to design a scalable overlay supporting the complexity of content-based networks, while satisfying the desirable properties large distributed systems should have. This dissertation presents the design of Mirinae, a new structured peer-to-peer overlay mesh based on the interests of peers. Due to the flexibility of virtual hypercube topology, Mirinae provides efficient event dissemination trees minimizing the participation of non-matching nodes. I also present a novel ID space transformation mechanism for balancing routing load of peers even with highly skewed data, which is typical of the real world. The evaluation demonstrates that Mirinae is able to achieve its goals of scalability, efficiency, and near-uniform load balancing. Mirinae can be used as a substrate for content-search and range query in other important distributed applications.