Implementing a relational database by means of specialzed hardware
ACM Transactions on Database Systems (TODS)
ACM Transactions on Database Systems (TODS)
Microprocessor Assisted Tuple Access, Decompression and Assembly for Statistical Database Systems
VLDB '82 Proceedings of the 8th International Conference on Very Large Data Bases
Statistical Databases: Characteristics, Problems, and some Solutions
VLDB '82 Proceedings of the 8th International Conference on Very Large Data Bases
| Hi-index | 0.00 |
This paper presents the design of a data base machine for supporting statistical data bases. The primary objective of this research is to introduce an architecture which performs efficiently in executing relational operations on fully transposed files. This objective is met by utilization of two functionally specialized subsystems: category and summary subsystems. The category subsystem exploits the parallel, content addressed search capabilities of associative memories (AM) while the summary subsystem employs a set of functionally equivalent processors suitable for evaluating statistical functions. In order to support processing of category attributes, the design of an extended associative memory has been considered and its features are presented. The most important feature of this AM stems from its ability to sort selected tuples of a relation, with respect to several attributes simultaneously and at the same time to produce an inverted list for each attribute. Sorting is performed by enumeration. The sorted inverted lists are used for efficient execution of relational projection and join operations.