CPISync

Set reconciliation synchronizes unordered data sets, that is to say that the order of information in the data sets being synchronized is not important - such as an appointment schedule, where the data can be easily re-ordered later according to the time of appointments (say). There are two approaches to CPISync possible, one when an upper bound on the number of differences $\overline{m}$ between the synchronizing data sets is known and the other when $\overline{m}$ is not known and has to be intelligently guessed. We call the former approach the ``Deterministic approach" and the latter ``Probabilistic approach". In the Probabilistic approach we iterate the Deterministic approach repeatedly by increasing $\overline{m}$ in each iteration until a satisfactory reconciliation results.