System-Wide Multiresolution
Robert L. Read, Donald S. Fussell, Avi Silberschatz · 1993
A major problem of current systems is that the quantity of data outstrips the system''s capacity. A number of existing systems address this problem by allowing quality to be traded for performance. We call such systems systems. This paper argues that ubiquitous, systematic support for for the tradeoff of quality for performance is essential to next-generation applications that access very large quantities of data. We demonstrate that current trends in computing technology will increase the importance of multiresolution in the future. We sketch a tentative plan for research into multiresolution database technology, and mention some research direction in other subdisciplines of Computer Science. Finally, we conclude by presenting an ambitious target application, the Geoscope, as a motivation of ubiquitous multiresolution.