Following decades of rapid and sustained advances in computing and communication technologies, we have reached a stage where it is becoming feasible to embed computing and communication functionality in the physical objects that surround us. For example, such functionality may be embedded into dishwashers, refrigerators, coffee machines, heating systems and even clothing and jewelry. Other, perhaps more mundane examples include PDAs, mobile phones, MP3 players, and car navigation systems. Many of these computationally enabled objects are small and mobile. As they are also able to communicate with their surroundings, e.g., via low-cost transceivers that allow them to spontaneously interconnect with other objects or via a cellular network, they will have access to the Internet and will be accessible from the Internet. Put differently, their data will be part of a global information space, and they will be able to exploit this space.

As a consequence, data is spread all over, and data synchronization and consistency become big issues. This is also true for data placement that comprises replication techniques as well as caching technology. This situation is further complicated by the fact that these data stores could be inherently heterogeneous in terms of data structuring, storage format, and access technology. In any case, it is worthwhile to rethink, whether the currently available solutions in data management still fit to these new scenarios, and how an appropriate data management technology that covers the whole spectrum would look like. It is exactly the focus of this seminar to discuss all these issues related to the management of mobile, ubiquitous, and pervasive data.

As a consequence, data is spread all over, which offers a host of new challenges to data management technology. In particular, data synchronization and consistency become substantial challenges. This is also true for data placement that comprises replication techniques as well as caching technology. This situation is further complicated by these data stores most likely being inherently heterogeneous in terms of data models, storage formats, and access technologies.

This development makes it worthwhile to rethink whether the currently available data management solutions are appropriate for these new scenarios, and how data management technologies that support the whole spectrum would look like. It is exactly the focus of this seminar to discuss issues related to the management of mobile, ubiquitous, and pervasive data.

The seminar brought together representatives from different communities (researchers, software vendors, and users) from different areas (mobile application, middleware, sensor systems, distributed systems, database systems) for joint, in-depth discussions of emerging data management challenges, key objectives being to identify research challenges and standardization needs, and to better understand open problems.

Workshop Itinerary

The seminar started with the following list of potential discussion points phrased as questions and given to the participants up front:

• How to synchronize the data and how to achieve consistency?
• How to integrate the data?
• Where to place the data and which technologies to use?
• How to manage, store, and access the data?
• What does a suitable processing model looks like?
• What kind of communication technology is needed?
• What about a platform approach?
• How to build these kinds of applications?
• What are the killer applications?
• How would a comprising technology for moving objects look like?

The five days were filled with discussions, workgroup meetings, and presentations. The organizers decided on purpose to leave substantial room for discussions and workgroup meetings, to enable the participants to cover new problems and topics that emerged as the seminar progressed and that were considered important for the development of the field. A good overview of the organization of the seminar is given in the Mind Map diagram that follows (prepared by Prof. Jano Moreira de Souza, UFRJ, Rio de Janeiro).

Workshop Résumé

At the end of the seminar, the participants categorized the issues that were considered during the presentations and discussions throughout the seminar week. The following list of general topics resulted. These topics capture well the breadth of the seminar. The topics were considered to be particularly interesting by the participants, primarily due to research challenges they embody or their potential relevance for future practice.

• Location-based and moving-objects-based applications
• Data/content integration, federation, and management
• Context management and context-aware services
• Data in sensor networks, data streams, and sensor fusion
• Modeling and querying of mobile and spatial databases
• Indexes for high update rates, including the indexing of the past, current, and near-future positions of moving objects
• Replication and caching
• P2P database middleware
• Privacy Issues

It is our belief that the seminar has improved the participants’ understanding of the seminar’s topic area in general and of the abovementioned topics in particular, has built new collaborations among the seminar participants, and will stimulate further collaborations among members of the different communities involved.