| In this dissertation, we explore three issues in interaction management and data access arising from the advent of mobile and distributed computing systems.; First, we consider the implications of mobility on data access . We adopt a broadcast-based strategy, in which large numbers of users can simultaneously access broadcast data. We design adaptive broadcast protocols to be employed by database servers to decide on the content of broad-casts dynamically, in response to client mobility and demand patterns. We also propose efficient retrieval strategies that may be employed by clients to download information from broadcasts.; Second, we expand our inquiry to consider transaction-oriented interactions in mobile environments. With the expansion of Web sites to support complex functions, a user interfaces with e-businesses through an interactive and multi-step process, which is often time-consuming. In mobile environments this “disconnection-reconnection-repeat work” cycle may cause web clients to incur substantial monetary as well as resource (such as battery power) costs. We propose a framework for reasoning over such interactions, as well as a protocol for “recovering” a user to an appropriate recent interaction state after such a failure. The objective is to minimize work that needs to be redone upon restart after failure.; Third, we consider the implications of distribution on web services interactions. The web services paradigm allows enterprises to share applications with customers, partners, and suppliers, regardless of hardware or software environment, using a set of standards-based protocols for intercommunication. In the context of web services, we explore two aspects: how a user can specify declaratively his service request; and how we can develop and execute “optimal” execution plans from that declarative request, where “optimal” can apply to any of a number of criteria, including finding the lowest-cost plan, or the one with the fastest response time. Specifically, we develop: (1) a framework for specifying service composition declaratively; and (2) a system architecture in which such specifications can be handled, including all necessary components, from gathering user input to generating an optimal execution plan from that input to actually executing plan and verifying the result. |
