Supporting Flat Relations by a Nested Relational Kernel
Generate an AI Snapshot to get a quick, structured summary of this paper.
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
A more general two-level rolution: The authors' databsse kernel offer6 hierarchical clnsteriug expmwed as nested (NF2) relationa Flat relations, resulting from logical database d&n, are then mapped to thir internal kernel interface.
Abstract
While a variety of sophisticated phy&al database design technique6 ha6 been deviled in research, only very limited capabilitie6 are available in practice. Thir ie mostly due to the fact, that designing intunal hsyoutr di!Tering from the loscal view b not transparent to the user. We present a more general two-level rolution: Our databsse kernel offer6 hierarchical clnsteriug expmwed as nested (NF2) relationa Flat relations, resulting from logical database d&n, are then mapped to thir internal kernel interface. We show, how the variour physical structuring approache6 cau be espre66ed in thi6 model. Phy6i’cd database de6ign for a flat relational front-end can then be described formally within the (NF2) relational model. The important sclpect of join 6upport h pursued by materializing 6ome join6 in ne6ted relation6 without any redundancy. S&&project-join quexiea on the logical rchema can be transformed to efficiently proce66ible intemd queries by applying algebraic optimization techniques, h~own e.g. from view optimization. P&minary performau ce evaluation6 are reported that were carried out on commerciay adable system6 and solicited our expectation6.
