|
ERDB Entity Related Database
|
|
ERDB Entity Related DataBase ERDB is an in-memory database []system from Celeram Ltd. ERDB is designed for mission-critical high-volume data applications and transaction processing. Unlike disk-based relational databases, ERDB's data is located in memory, thus removing the bottleneck of disk I/O. ERDB differs from most in-memory databases in that the data is not stored in relational tables, but in linked lists, using binary tree technology. This design offers great advantages in terms of perdictability, access speed and space efficiency. Data can be queried using industry-standard tools such as SQL, ODBC and JDBC or by using the product's native API (Application Programming Interface) History ERDB was originally developed at Automated Technology Associates of Indianapolis, USA by a team led by David Layden in 1984. Up until then, relational databases had been used in-memory to obtain the necessary speed, but there were weaknesses inherent in the nature of relational technology. Early adopters of the technology were manufacturing companies such as Ford, General Motors and Bayer Chemicals but later telecom companies such as Orange and Vodafone began to adopt the product. Product Information ERDB is a Fourth Normal Form database. All ERDB relationships (or functional query assumptions) are predefined and built into the application. Thus a “direct” connection, not an implied one, exists between relationships. It is ERDB’s predefined threaded query structure that, in fact, permits response times to remain predictable even as the database application grows in size and complexity. ERDB’s innovative application of linked lists and indirect references provides the most direct path to the data and the best utilization of storage space. By maintaining the direct connection between relationships, no normalization of data is required as is no third normal form table structures. In addition, by implementing this efficient, compact mechanism for managing concatenated keys, the need for duplicate data is eliminated as are the problems associated with sparse matrix data structures. ERDB combines high performance with highly predictable response-a must for some applications. 
|
| < Prev | Next > |
|---|
Main Menu
| Home |
| Articles |
| Search |
| Submit Article |
| Contact Us |
| Donate to Wikibin |
