CSRG-585, Middleware Systems Research Group, University of Toronto, October 2007.
This paper introduces to the publish/subscribe model the ability to uniformly access data produced in the past and future. The new model can filter, aggregate, correlate and project any combination of historic and future data. A flexible architecture is proposed consisting of distributed and replicated data repositories that can be provisioned in ways to tradeoff availability, storage overhead, query overhead, query delay, load distribution, parallelism, redundancy and locality. Evaluations in a distributed testbed show that different provisioning policies perform better with read or write heavy workloads. In addition, a dynamic query routing algorithm is used to optimize the location where sub-queries are evaluated, yielding traffic reductions of up to 72%.