On Real-Time Top k Querying for Mobile Services

Abstract

Mobile services offering multi-feature query capabilities must meet tough response time requirements to gain customer acceptance. The top-k query model is a popular candidate to implement such services. Focusing on a central server architecture we present a new algorithm called SR-Combine that closely self-adapts itself to particular cost ratios in such environments. SR-Combine optimizes both the object accesses and the query run-times. We perform a series of synthetical benchmarks to verify the superiority of SR-Combine over existing algorithms. In order to assess whether it can meet the stated response time requirements for mobile access, we propose a psychologically founded model. It turns out that for a wide range of practical cases SR-Combine can satisfy these goals. Where this isn't yet the case, we show up ways how to get there systematically. Thus with SR-Combine a breakthrough in real-time capabilities of top-k querying for mobile services is in sight now.

Efficient Distributed Skylining for Web Information Systems

“…using a suitable index) will of course complement the total runtime improvement by reducing the overall number of accesses. Therefore minimizing the number of necessary object accesses and thus also the overall query runtimes is tantamount to build practical systems (with real-time constraints) [1]. Prototypical Web information systems of that kind are e.g.…”

“…Please note that our heuristic 1 will not affect the abstract order of complexity from our previously stated optimality results, because the maximum improvement factor over the round robin strategy can only be the number of lists (n). But, given the rather expensive costs of object accesses over the Internet even small numbers of accesses saved will improve the overall run-time behavior like shown in [5] or [1]. Thus, also improvements taking only constant factors off the algorithm's complexity should be employed towards meeting real-time constraints.…”

“…Recent research on web-based information systems has focused on employing middleware algorithms, where users had to specify weightings for each aspect of their query and a central compensation function was used to find the best matching objects [7], [1]. The lack of expressiveness of this 'top k' query model, however, has first been addressed by [8] and with the growing incorporation of user preferences into database systems [6], [10] and information services [22] the limitations of the entire model became more and more obvious.…”

A survey on representation, composition and application of preferences in database systems

“…For example, the N RA algorithm is appropriate when random accesses are expensive or impossible and so only sorted accesses are employed, while the C A algorithm is appropriate when random accesses are expensive relative to sorted accesses [Fagin et al 2001]. Balke et al [2002] propose the SR-Combine algorithm for top-k query processing in mobile environments. All the aforesaid approaches focus on constructing an aggregate ranking by combining a set of rankings that contain the same set of tuples.…”

Multi-objective Query Processing for Database Systems

“…Please note, that in this case all database objects are ranked into a (non-strict) total ordering. Besides central index approaches like k-nearest-neighbor searches, the algorithm using a threshold condition has been proved to be most efficient for applications [20], [11], [21], [6], [2] and has even been shown to be instance-optimal [10]. To be self-contained we present the basic algorithm below.…”