Show simple item record

Trie Hashing with Controlled Load .

dc.contributor.authorLitwin, Witolden_US
dc.contributor.authorRoussopoulos, N.en_US
dc.contributor.authorLevy, G.en_US
dc.contributor.authorWang, H.en_US
dc.description.abstractTrie hashing is an access methods to primary key ordered dynamic files. The key address is computed through a trie. Key search needs usually one disk access since the trie may be in core and needs two accesses for very large files, when the trie has to be on the disk. We present a new variant of the method that allows to set up an arbitrary load factor for ordered insertions. In particular, one may create compact files, loaded up to 100%. We show that the capabilities of trie hashing make the method preferable to a B-tree by most of criteria that motivated the latter method supremacy over the database world.en_US
dc.format.extent1445929 bytes
dc.relation.ispartofseriesISR; TR 1989-13en_US
dc.titleTrie Hashing with Controlled Load .en_US
dc.typeTechnical Reporten_US

Files in this item


This item appears in the following Collection(s)

Show simple item record