Related Experiment Videos
A new concept of sequence data distribution on wide area networks
K Heumann1, D George, H W Mewes
1Martinsried Institute for Protein Sequences (MIPS), Max-Planck-Institute for Biochemistry, Germany.
Computer Applications in the Biosciences : CABIOS
|September 1, 1994
Summary
This study presents a network system for synchronizing remote databases using incremental updates. It ensures consistent, uninterrupted access to distributed resources despite network instability.
Area of Science:
- Computer Science
- Distributed Systems
- Database Management
Background:
- Centralized access to computerized resources is often undesirable due to hardware, network bandwidth, and stability limitations.
- A distributed network model is needed to overcome the challenges of remote database synchronization.
Purpose of the Study:
- To develop a network-based system for synchronizing remote sequence database access sites.
- To ensure consistent and uninterrupted access to distributed resources.
Main Methods:
- Developed a network model to distribute access across remote computer centers.
- Investigated distributed transaction protocols to guarantee node consistency.
- Implemented a layered software architecture making protocols transparent.
- Utilized Remote Procedure Calls (RPCs) for component communication.
Main Results:
- A prototype system was implemented to synchronize distributed copies of the PIR-International Protein Sequence Database.
- The system successfully compensated for unstable network links, ensuring uninterrupted resource access.
- The layered architecture facilitated transparency for database systems and the network.
Conclusions:
- The developed network system effectively synchronizes remote databases using an incremental update mechanism.
- The system provides a robust solution for distributed data access, ensuring consistency and availability.
- This approach is suitable for managing heterogeneous, geographically dispersed data resources.