Abstract not found

Abstract. As low-level architectural support for context-aware computing matures, we are ready to explore more general and powerful means of accessing context data. Information required by a context-aware application may be partitioned by any number of physical, organizational, or privacy boundaries. This suggests the need for mechanisms by which applications can issue context-sensitive queries without having to explicitly manage the complex storage layout and access policies of the underlying data. To address this need, we have developed liquid, a prototype query service that supports distributed, continuous query processing of context data. This paper articulates the current need for such systems, describes the design of the liquid system, and presents both a room-awareness application and notification service demonstrating its functionality. 1

Abstract — Distributed stream-based applications, such as continuous query systems, have network scale and time characteristics that challenge traditional distributed query optimization. The optimization sub-problems of plan generation and service placement should be integrated to meet these challenges. These tasks have typically been treated as independent sub-problems because of the complexity of their integration. We suggest cost spaces as one way to mitigate this complexity. We further consider how cost spaces can be used to allow tractable multi-query optimization. I.

Introduction 1.1 A Vision for the Future Internet Today's Internet is almost unrecognizable compared to even a few years ago. The WWW has exploded in a way no one had foreseen. The Web browser has emerged as a universal client and universal user interface for accessing not only static Internet content, but also services that have migrated to the Web: online commerce (E-trade), entertainment (CNN, ESPNet), reservations services (Internet Travel Network), search engines (HotBot, AltaVista), reference materials (Encyclopedia Britannica, Webster's Online), interactive maps (MapQuest), and a plethora of others. Despite these advances, the Internet, and the Web in particular, remain in many ways extremely primitive. The down side of the "universal client" effect is that servers (and services) now assume that the hardware and software platforms on which those clients are hosted and the networks to which they're connected are also

We consider the problem of query optimization in distributed data stream systems where multiple continuous queries may be executing simultaneously. In order to achieve the best performance, query planning (such as join ordering) must be considered in conjunction with deployment planning (e.g., assigning operators to physical nodes). In our scenario, the large number of network nodes, query operators, and opportunities for operator sharing between queries means that brute force and traditional techniques are too expensive. We propose two algorithms- the Bottom-Up algorithm and the Top-Down algorithm, which utilize hierarchical network partitions to provide scalable query optimization. We present analysis that establishes the bounds on the search-space and sub-optimality achieved by our algorithms. Finally, through simulations and experiments using a prototype deployed on Emulab [1] we demonstrate the effectiveness of our algorithms. The Top-Down algorithm, for instance, was able to achieve, on an average, solutions that were sub-optimal by only 10 % while considering less than 1 % of the search space. 1.

Abstract—This paper addresses the problem of optimizing multiple distributed stream queries that are executing simultaneously in distributed data stream systems. We argue that the static query optimization approach of “plan, then deployment ” is inadequate for handling distributed queries involving multiple streams and node dynamics faced in distributed data stream systems and applications. Thus, the selection of an optimal execution plan in such dynamic and networked computing systems must consider operator ordering, reuse, network placement, and search space reduction. We propose to use hierarchical network partitions to exploit various opportunities for operator-level reuse while utilizing network characteristics to maintain a manageable search space during query planning and deployment. We develop top-down, bottom-up, and hybrid algorithms for exploiting operator-level reuse through hierarchical network partitions. Formal analysis is presented to establish the bounds on the search space and suboptimality of our algorithms. We have implemented our algorithms in the IFLOW [1] system, an adaptive distributed stream management system. Through simulations and experiments using a prototype deployed on Emulab [2], we demonstrate the effectiveness of our framework and our algorithms. Index Terms—Computer-communication networks, distributed systems, distributed databases, distributed applications, database management, systems, query processing. Ç 1