| | |
Summary: Internet-scale storage systems under churn
-- A study of the steady-state using Markov models
Anwitaman Datta, Karl Aberer
Ecole Polytechnique F´ed´erale de Lausanne (EPFL)
School of Computer and Communication Sciences (I&C)
CH-1015 Lausanne, Switzerland
Anwitaman.Datta@epfl.ch, Karl.Aberer@epfl.ch
Abstract
Content storage in a distributed collaborative environ-
ment uses redundancy for better resilience and thus pro-
vides good availability and durability. In a peer-to-peer en-
vironment, where peers continuously leave and rejoin the
network, various lazy strategies can be employed to main-
tain a minimal redundancy of stored content in the system.
Existing static resilience analyses fail to capture in detail
the system's behavior over time, particularly the probabil-
ity mass function of the actual available redundancy, since
it ignores the crucial interplay between churn and main-
tenance operations, and looks only at the average system
property. We perform a Markovian time-evolution analysis
|
