Spurring Adoption of DHTs with OpenHash, a Public DHT Service

Abstract

The past three years have seen intense research into Distributed Hash Tables (DHTs): both into algorithms for building them, and into applications built atop them. These applications have spanned a strikingly wide range, including file systems [1–3], event notification [4], content distribution [5], e-mail delivery [6], indirection services [7,8], web caches [9], and relational query processors [10]. While this set of applications is impressively diverse, the vast majority of application building is done by a small community of DHT researchers. If DHTs are to have a positive impact on the design of distributed applications used by real users outside this research community, we believe that the community of DHT-based application developers should be as broad as possible. Why, then, has this community of developers remained narrow? First, keeping a research prototype of a DHT running continually requires effort, and experience with DHT code. Second, significant testbed resources are required to deploy and test DHT-based applications. A hacker can download the code for Chord, but she cannot run that code alone; recall that only a tiny fraction of would-be developers has access to a testbed infrastructure like PlanetLab [11]. Consequently, most application developers would turn to ad hoc application-specific solutions rather than attempt to use a DHT.