A case for redundant arrays of inexpensive disks (RAID)
SIGMOD '88 Proceedings of the 1988 ACM SIGMOD international conference on Management of data
Efficient organization of semantic databases
3rd International Conference, FODO 1989 on Foundations of Data Organization and Algorithms
Proceedings of the world transputer user group (WOTUG) conference on Transputing '91
A file structure for semantic databases
Information Systems
Interconnection Networks for Parallel and Distributed Processing
Interconnection Networks for Parallel and Distributed Processing
SM3: A Dynamically Partitionable Multicomputer System with Switchable Main Memory Modules
Proceedings of the First International Conference on Data Engineering
Relational Algebra Machine GRACE
Proceedings of RIMS Symposium on Software Science and Engineering
The architecture and the related control problems of a transputer-based highly parallel database computer
Hi-index | 0.00 |
Parallel computing is a promising way to achieve high performance in a database system. The disk access speed has been a well known bottleneck for database machines, and the data intensive nature and the random communication patterns of databases make the interconnection network in a database machine difficult to design.This article describes the design of a highly parallel, high throughput database machine based on the new T9000 transputer family and a large number of relatively inexpensive disks. The database machine is designed to achieve a throughput of more than 10,000 elementary transactions per second and a total size in the range of terabytes. The Semantic Binary Database Model is employed by the system. Besides its logical aspects, this model's implementation facilitates the utilization of the parallel processing power and improves the reliability of the system.