Domus - An Architecture for Cluster-oriented Distributed Hash Tables
José Rufino, António Manuel Silva Pina, Albano Alves, José Exposto · 2005
Abstract. This paper presents a high level description of Domus, an architecture for cluster-oriented Distributed Hash Tables. As a data management layer it supports the concurrent execution of multiple and heterogeneous DHTs, that may be simultaneously accessed by different distributed/parallel client applications. At system level, a global balancement mechanism allows for the (re)distribution of each DHT over cluster nodes, based on the monitoring of the overall cluster resources, including memory, CPUs, storage and network. Two basic units of balancement are supported: vnodes, a coarse-grain unit, and partitions, a fine-grain unit. The design, at each cluster node, and for each DHT, also takes advantage of the strict separation of object lookup and storage. Lookup follows a distributed strategy that benefits from the joint analysis of multiple partition-specific routing information (from a routing overlay that interconnects all the partitions), to shorten routing paths. Storage is accomplished by supporting different data repositories according to the specificity and requirements of each DHT. 1