Personal Data Servers

Project Background

An increasing amount of personal data is automatically gathered and stored on servers by administrations, hospitals, insurance companies, etc. Citizens themselves often count on internet companies to store their data and make them reliable and highly available through the internet. However, these benefits must be weighed against privacy risks incurred by centralization. The project suggests a radically different way of considering the management of personal data. By embedding a full-fledged Personal Data Server in mobile devices, user control of how his/her sensitive data is shared by others (by whom, for how long, according to which rule, for which purpose) can be fully reestablished and convincingly enforced. To give sense to this vision, Personal Data Servers must be able to interoperate with external servers and must provide traditional database services like durability, availability, query facilities, transactions.

Current Milestones

1. Rudimentary peer-to-peer replication developed
2. Local (individual PDS node) Query Processing developed
3. Publish-Subscribe model deployed with broker network comprising broker and back-end DB

Tentative RoadMap

1. Partitioned replication development
2. Global Query Processing
3. Geographical PDS nodes partitioning and efficient routing based on the broker network location
4. PKI Protection of query results

PlanetLab Simulations

Since personal databases deployed on a variety of mobile devices could be connected to Internet in a variety of ways we plan to test the deployment of PDS publish-subscribe model on the real Internet. We are conducting internet-scale tests on the PlanetLab testbed to verify the protocols design scalability on real Internet with geographically dispersed PDS nodes to test QoS of replication and query processing facilities as well as the scalability of the broker server network in the presence of real-world delays, packet losses and latency of the links.

Relevant Publications (Related Work)

1. Personal Data Servers: a Vision Paper
2. PicoDBMS: Scaling down database techniques for the smartcard
3. Secure and Customizable Data Management for Automotive Systems: A Feasibility Study
4. Chip-Secured Data Access: Confidential Data on Untrusted Servers
5. The Ubiquitous DBMS
6. Mobile Databases: a Selection of Open Issues and Research Directions
7. Conception of Data Base Management System in USB Smart Card Flash Memory: Application for the Cancer Pathology of Medical Information Systems
8. Stream Processing Health Card Application
9. Hippocratic Databases
10. Data management issues in mobile and peer-to-peer environments
11. TinyDB: An Acquisitional Query Processing System for Sensor Networks
12. Integrating Databases with Publish/Subscribe

Useful links:

1. Embedding an SQL Database with SQLite
2. Using Java DB in Desktop Applications
3. Working with the Java DB (Derby) Database
4. Embedded Derby (Java DB)

Download preliminary PDS software Java package (alpha stage) - coming soon