MRDBM – A multi‐user relational database machine
Y.K. Liu, Keh‐Yih Su
Abstract
Y.K. Liu, Keh‐Yih Su
Abstract
In order to satisfy the future demands on database processing, a new database machine termed Multi‐user Relational Database Machine (MRDBM) is proposed in this paper. It executes database operations in a parallel manner by applying distributed multiple processing units. However, it differs from other database machines using the same approach in three ways: (1) data pages rather than whole relations (or buckets) are used as the processing entity to increase processing parallelism; (2) data pages are embedded with instruction codes to reduce control overhead; (3) heterogeneous architecture is adopted to make the processing of complex database operations like sort and join more efficient. As a result of these modifications, MRDBM has better performance especially in the multi‐user environment. In this paper, we will describe how the system is designed and the current development of MRDBM.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In order to satisfy the future demands on database processing, a new database machine termed Multi‐user Relational Database Machine (MRDBM) is proposed in this paper. It executes database operations in a parallel manner by applying distributed multiple processing units. However, it differs from other database machines using the same approach in three ways: (1) data pages rather than whole relations (or buckets) are used as the processing entity to increase processing parallelism; (2) data pages are embedded with instruction codes to reduce control overhead; (3) heterogeneous architecture is adopted to make the processing of complex database operations like sort and join more efficient. As a result of these modifications, MRDBM has better performance especially in the multi‐user environment. In this paper, we will describe how the system is designed and the current development of MRDBM.
Key concepts: Computer science, View, Relational database, Database, sort, Relational database management system, Overhead (engineering), Database tuning