First, we introduce the problem of managing and retrieving data in distributed
systems, compare three basic approaches for this, and show that
some of them do not scale well, even though they are Peer-to-Peer approaches.
As a result, we introduce the promising concept of the Distributed Hash Table
(DHT) for designing and deploying highly scalable distributed systems.
In this chapter, we focus only on the basic properties and mechanisms of
Distributed Hash Tables; specifics of certain DHT approaches are presented
in the next chapter.
The remainder of this chapter is organized as follows. After discussing
in Section 7.1 general concepts for distributed management and retrieval of data in Peer-to-Peer systems, the subsequent sections introduce Distributed Hash Tables, in particular their fundamentals (Section 7.2), the concept of content-based routing (Section 7.3), and DHT interfaces (Section 7.4). The next chapter presents specific algorithms of popular DHT approaches, e.g., how to organize the address space, and how routing in a Distributed Hash Table is performed. In Chapter 9, we discuss aspects of reliability and loadbalancing in Distributed Hash Tables.