Optimization techniques for data-intensive decision flows

For an enterprise to take advantage of the opportunities afforded by electronic commerce it must be able to make decisions about business transactions in near-real-time. In the coming era of segment-of-one marketing, these decisions will be quite intricate, so that customer treatments can be highly personalized, reflecting customer preferences, the customer's history with the enterprise, and targeted business objectives. This paper describes a paradigm called "decision flows" for specifying a form of incremental decision-making that can combine diverse business factors in near-real-time. This paper introduces and empirically analyzes a variety of optimization strategies for decision flows that are "data-intensive", i.e. that involve many database queries. A primary focus is on the use of parallelism and eagerness (a.k.a. speculative execution) to minimize work and/or reduce response time. A family of optimization techniques is developed, including algorithms and heuristics for scheduling tasks of the decision flow. Using a prototype execution engine the techniques are compared and analyzed in connection with decision-making applications having differing characteristics.