Skip to main content
European Commission logo print header

DBGlobe: A Data-centric Approach to Global Computing

Objective

On one view, global computing can be seen as a database problem: how to design, build and analyse systems that manage large amount of data. However, the traditional database approach of storing data of interest in monolithic database management systems becomes obsolete in such environments. In current database research, data are relatively homogeneous, exhibit a small degree of distribution (just a few network sites) are passive in that they remain unchanged unless explicitly updated. All these assumptions do not hold in the global computing world. This creates the need for new theoretical foundations in all aspects of data management: modelling, storage and querying.

OBJECTIVES
DBGlobe aims at revolutionizing the way we think of databases, as the mundane task of processing static data in monolithic database management systems. It broadens database management research focus to attack the issues of mobility, autonomy, incomplete information, scale, and adaptability that arise in dynamic environments.

DESCRIPTION OF WORK
DBGlobe adapts a data-centric approach to the design and analysis of dynamic environments of autonomous and mobile entities by considering each mobile entity as a primary data store and a mini-server that protects and encapsulates access to its data. Such networks of mobile entities constitute ad-hoc distributed database systems of unprecedented scale. Besides these "walking" databases of mobile entities, in DBGlobe, meta-information and services related to them are maintained in dedicated data stores, called InfoStations, dispersed throughout the stationary network.

InfoStations have two components:
i) DataStores that hold metadata about mobile entities, called mobile entity profiles, with information such as the location and type of the entities, and;
ii) DataHandlers that provide services for and about the mobile entities. Such networks of InfoStations create a backbone of metadata information and services and provide the means for reasoning about and querying the behaviour and state of the autonomous mobile entities.

The collection of databases on mobile entities that exist around a specific context (e.g. location or user) form a data sharing community that we call an ad-hoc database. DBGlobe is designed to facilitate the seamless creation and operation of such ad-hoc databases.
In particular, within the DBGlobe project, we will develop novel data management mechanisms along the following key topics:
1. System Architectures: There is no centralized database server, instead, each mobile object constitutes a database of each own. In this context, our specific research objectives are:
- to define what is the appropriate metadata information to describe mobile entities;
- to design distribution and replication protocols for DataHolders and the DataHandlers;
- to make the architectures dynamically configurable and extensible and to achieve fault-tolerance and availability;
2. Co-ordination/Data Delivery: An important component of DBGlobe is data delivery among the components that participate in the system: a) the mobile entities, b) the InfoStations and c) the users. Current database servers support a pull-based mechanism of data delivery: data are sent to a client after having been explicitly requested. This request-response type of data delivery is not appropriate for all interactions in the DBGlobe system.

Our objective is to derive adaptive data delivery mechanisms that will combine: i) push (transmission of data without an explicit request) and pull, ii) periodic and aperiodic and iii) multicast and unicast delivery. In addition, we intend to model the co-ordination of the mobile entities using workflow management. We also intend to explore techniques that have been used in the multi-agent community;
3. Querying: In DBGlobe, data exchange and computation occur in the background in response to cues or queries from users. Querying is performed on a multitude of databases (networked mobile processing entities and their data) rather than on a centralized device or system. New forms of query languages and query management paradigms need to be developed.

Our objective is to derive new query language modelling abstractions that will:
- include a knowledge acquisition component;
- incorporate filtering, since some of the devices may continuously produce and transmit data;
- Be context-aware, to deal with the scale and complexity of the environment.
In addition, we intend to work towards query execution models for such dynamic environments of numerous processing entities;

4. Simulation and Proof-of-Concept Prototype: To verify our model we intend to: a) build a simulator for such dynamic environments and use it to test our protocols and b) implement a proof-of-concept prototype. Our simulator will model mobile entities and their interactions. It will be incrementally extended to model the creation of ad-hoc databases, delivery of data, co-ordination and querying.

Call for proposal

Data not available

Coordinator

UNIVERSITY OF IOANNINA
EU contribution
No data
Address
PANEPISTIMIOUPOLI IOANNINON
45110 IOANNINA
Greece

See on map

Total cost
No data

Participants (5)