We consider queries which originate from a mobile unit and whose result depends on the location of the user who initiates the query. Example of such a query is How many people are living in the region I am currently in?” We execute such queries based on location-dependent data involved in their processing. We build concept hierarchies based on the location data. These hierarchies define mapping among different granularities of locations. One such hierarchy is to generate domain knowledge about the cities that belong to a state. The hierarchies are used as distributed directories to assist in finding the database or relation that contains the values of the location-dependent attribute in a particular location. We extend concept hierarchies to include spatial indexes on the location-dependent attributes. Finally, we discuss how to partition and replicate relations based on the location to process the queries efficiently. We briefly discuss the implementation issues.

The performance of network and communication software is a major concern for making the electronic commerce applications in a distributed environment a success. The quality of service in electronic commerce can generically be measured by convenience, privacy/security, response time, throughput, reliability, timeliness, accuracy, and precision. We present the quality of service parameters, software architecture used in e-commerce, experimental data about transaction processing in the internet, characteristics of digital library databases used in e-commerce and communication measurements for such data. We present a summary of e-commerce companies and their status and give an example of electronic trading as an application.

The performance of OLAP queries can be improved drastically if the warehouse data is properly selected and indexed. The problems of selecting and materializing views and indexing data have been studied extensively in the data warehousing environment. On the other hand, data partitioning can also greatly increase the performance of queries. Data partitioning has advantage over data selection and indexing since the former one does not require additional storage requirement. In this paper,we show that it is beneficial to integrate the data partitioning and indexing (join indexes)techniques for improving the performance of data warehousing queries.We present a data warehouse tuning strategy, called PartJoin, that decomposes the fact and dimension tables of a star schema and then selects join indexes. This solution takes advantage of these two techniques, i.e., data partitioning and indexing. Finally,we present the results of an experimental evaluation that demonstrates the effectiveness of our strategy in reducing the query processing cost and providing an economical utilisation of the storage space.

Setting up experimental networks of a sufficient size is a crucial element for the development of communication services. Unfortunately, the required equipment, like routers and hosts, is expensive and its availability is limited. On the other hand, simulations often lack interoperability to real systems and scalability, which limits the scope and the validity of their results. Therefore, an intermediate approach between these two alternatives that allows for setting up testbeds on a cluster of computers is needed. This paper presents an intermediate approach based on the emulation of IP routers and evaluates the concept. In a first set of experiments the impact of various parameters on the packet delay was investigated, while further experiments compare the performance of differentiated services run on the network emulator with the results obtained by the well known network simulator ns.

To avoid a congestion collapse, network flows should adjust their sending rates. Adaptive flows adjust the rate, while unresponsive flows do not respond to congestion and keep sending packets. Unresponsive flows waste resources by taking their share of the upstream links of a domain and dropping packets later when the downstream links are congested We use network tomography-an edge-to-edge mechanism to infer per-link internal characteristics of a domain-to identify unresponsive flows that cause packet drops in other flows. We have designed an algorithm to dynamically regulate unresponsive flows. The congestion control algorithm is evaluated using both adaptive and unresponsive flows, with sending rates as high as four times of the bottleneck bandwidth, and in presence of short and long-lived background traffic.

Data warehouse [2, 4, 5, 6] is an integrated repository derived from multiple source (operational and legacy) databases. The data warehouse is created by either replicating the different source data or transforming them to new representation. This process involves reading, cleaning, aggregating and storing the data in the warehouse model. The software tools are used to access the warehouse for strategic analysis, decision-making, marketing types of applications. It can be used for inventory control of shelf stock in many departmental stores.

Multimedia data are usually very large. Fragmentation of multimedia data units is inevitable when they are transmitted through networks. Active networks are becoming popular, and active technologies are being applied to various interesting problems. When applying active technologies to multimedia data, however, the problem of fragmenting large packets still exists. Furthermore, new issues emerge when active capsules are fragmented. In this paper, we propose a new fragmentation scheme which addresses the unique needs of active networks and which utilizes the special properties of active networks. We propose an algorithm to fragment the data at the transport layer, which can minimize the overhead. Preliminary experimental results show that the scheme works well under realistic scenarios, with an overhead of less than 5%

We present the design and prototype of GnuStream, a peer- to-peer (P2P) and receiver-driven media streaming system. GnuStream is built on top of Gnutella, and it integrates dynamic peer location and streaming capacity aggregation. Each GnuStream streaming session is controlled by the receiver peer and involves a dynamic set of peer senders instead of one fixed sender. The receiver aggregates streaming bandwidth from the multiple senders, achieving load distribution and fast reaction to sender capacity and on/off-line status changes. The effectiveness of GnuStream is demonstrated by our experiments with its prototype, which serves as the basis for real-world development and evaluation of resilient P2P media streaming services.

In active network environments, the data packets can carry active programs to enable specialized processing on them. We quantify the effectiveness of general capsule programs vs. specialized processing functions for multimedia data through four experiments. These experiments deal with the comparison of Java vs. C implementation of a MPEG video decoder, the identification of the time-consuming modules in a Java MPEG video decoder, the effectiveness of combining Java methods and native methods, and the size of the bytecode for each module in a Java MPEG video decoder. We found that a Java MPEG decoder can be 2.6 times to 10 times slower than an equivalent C implementation. We identified the Huffmann decoding module as the most time-consuming module. We also found that the Native Method Interface (NMI) is complex and not efficient enough for use in active routers and the size of the bytecodes for most of the modules is too big to fit into a single packet even after compression. We draw certain conclusions about the trade-off between the general programming model and the specialized functions provided by the router for the active capsules

Recent software systems usually feature an automated failure reporting component, with which a huge number of failures are collected from software end-users. With a proper support of failure indexing, which identifies failures due to the same fault, the collected failure data can help developers prioritize failure diagnosis, among other utilities of the failure data. Since crashing failures can be effectively indexed by program crashing venues, current practice has seen great success in prioritizing crashing failures. A recent study of bug characteristics indicates that as excellent memory checking tools are widely adopted, semantic bugs and the resulting noncrashing failures have become dominant. Unfortunately, the problem of how to index non-crashing failures has not been seriously studied before. In previous study, two techniques have been proposed to index noncrashing failures, and they are T-Proximity and R-Proximity. However, as T-Proximity indexes failures by the profile of the entire execution, it is generally not effective because most information in the profile is fault-irrelevant. On the other hand, although R-Proximity is more effective than T-Proximity, it relies on a sufficient number of correct executions that may not be available in practice. In this paper, we propose a dynamic slicing-based approach, which does not require any correct executions, and is comparably effective as R-Proximity. A detailed case study with gzip is reported, which clearly strates the advantages of the proposed approach.

We have designed an adaptable network architecture, called ADNET, which provides mechanisms to allow an application to adapt to resource constraints to achieve improved QoS. In our experiments we compare three schemes (IP fragmentation, ACTP fragmentation with or without active program) of video transmissions. We find QoS is improved in the ACTP scheme with active programs. Our design aims to unify different QoS control mechanisms together to provide a wide range of network services to all users and meet their specific needs