Abstract
Structured peer-to-peer (P2P) networks have become an important
alternative to the classic client-server architecture. Participating
peers can join and leave the system at arbitrary times, a process which
is known as churn. Many recent studies revealed that churn is one of
the main problems faced by any Distributed Hash Table (DHT). As a
countermeasure most DHT protocols have to invest a significant amount
of overhead to maintain both the structure of the overlay and the
redundancy of the stored data.
In order to automatically adapt this maintenance overhead and other
parameters of the DHT to the changes in the overlay, an estimation of
the current churn rate is required. In this paper we discuss different
possibilities of how to estimate the current churn rate in the system.
In particular, we show how to obtain a robust estimate which is
independent of the implementation details of the DHT. We also
investigate the trade-offs between accuracy, overhead, and
responsiveness to changes.
Users
Please
log in to take part in the discussion (add own reviews or comments).