Design and Implementation of a centralized Backup and Restore Facility for distributed Systems

Introduction

Within the GAL-project a Multi-Services-Home-Platform (MSHP) is developed and it is planned for the future to install several instances of this platform in realistic environments. At the current state of the project, this platform is far from being reliable and stable and there is a high risk for a loss of data. Because of this, a local back-up-component already has been implemented and the transfer of back-ups to a central server is provided. The current system is limited to a local (at each MSHP) configuration and there is no strategy or system for handling backups and restoring data.

The basic approach for achieving data-security in the GAL-project is the following: all user data is only stored on the users’ MSHP. Data that leaves the MSHP is only allowed to be transferred in an encrypted way. Furthermore, backups have to be encrypted with a user-specific key or certificate, to assure that the backups can only be restored on the users’ MSHP.

Task Description

Within this thesis, a centralized backup and restore facility for several distributed MSHPs, operated via a remote Graphical-User-Interface (GUI), is to be designed and implemented; several sub-tasks can be identified:

Encryption

As encryption of all data leaving the MSHP is a crucial part, the first task should start with an overview and an analysis of recent existing encryption methods and their suitability for the desired application. An appropriate solution should include both, encryption and authentication. The used encryption-algorithms should be integrated in a modular way to easily allow a replacement in case of a possible corruption of the algorithms in the future.

Integrity Check

A separate integrity check should provide a mechanism for detecting the correct transfer and storage of the backup-files. It should be implemented for both directions: For backup-files transferred form an individual MSHP to a central storage and the other way around.

Local restore functionality

A separate integrity check should provide a mechanism for detecting the correct transfer and storage of the backup-files. It should be implemented for both directions: For backup-files transferred form an individual MSHP to a central storage and the other way around.

Central Configuration Management

The central facility should enable remote management and configuration of the distributed backup processes. Individual configurations for each installation should be supported as well as general configurations for all MSHPs.

Central Management of local and central Backups

The central management should include an overview of local backups (on each MSHP) and remote backups (stored in one or more central storages). Through a web-based GUI, the simple restore of individual Backups (stored remote or local) can be initiated.

During a concluding evaluation, the functionality of the backup and restore facility has to be shown and possible limitations or bottlenecks are to be identified.

• Fernwartung von Multi-Services-Home-Plattformen
• Monitoring eines Body Area Netzwerks