How to assign votes in a distributed system
Journal of the ACM (JACM)
Multi-disk management algorithms
SIGMETRICS '87 Proceedings of the 1987 ACM SIGMETRICS conference on Measurement and modeling of computer systems
Emerging trends in the WWW user population
Communications of the ACM
The dangers of replication and a solution
SIGMOD '96 Proceedings of the 1996 ACM SIGMOD international conference on Management of data
The SIFT information dissemination system
ACM Transactions on Database Systems (TODS)
Consistency and locking for distributing updates to web servers using a file system
ACM SIGMETRICS Performance Evaluation Review
Bistro: a framework for building scalable wide-area Upload applications
ACM SIGMETRICS Performance Evaluation Review
The Gamma Database Machine Project
IEEE Transactions on Knowledge and Data Engineering
The Grid Protocol: A High Performance Scheme for Maintaining Replicated Data
Proceedings of the Sixth International Conference on Data Engineering
A Fault-Tolerant Algorithm for Replicated Data Management
Proceedings of the Eighth International Conference on Data Engineering
Duplicate Removal in Information System Dissemination
VLDB '95 Proceedings of the 21th International Conference on Very Large Data Bases
Weighted voting for replicated data
SOSP '79 Proceedings of the seventh ACM symposium on Operating systems principles
| Hi-index | 0.00 |
Abstract: This experience paper describes newsblaster.com, a commercial document management system that facilitates exchange of news worthy documents between two user communities: a) editor of newspapers and magazines who seek to publish articles, and b) member companies who want to disseminate information such as new product offerings, product recalls, alliances with other companies, etc. We focus on the Internet connection of a single node web server as the system bottleneck and discuss three alternative approaches to distribute the application across multiple nodes. The first, termed vertical, analyzes the semantics of the application and distributes it across multiple nodes based on its community of users and how they access the system. The second, termed horizontal, declusters the application's data across multiple nodes with the objective to distribute client requests evenly across these nodes. Hybrid, our third approach, considers a horizontal partitioning of services identified by a vertical partitioning. We describe the application of each approach to newsblaster.