On Sunday 29 October 2006 18:47, Joe Wilson wrote:
> The last test that simulated pre-sorting 2 million rows
> in index order prior to insert may show optimal insert speed
> since it's only appending pages to the table and the index,
> but it may not be realistic, since you may not have sufficient
--On Friday 27 October 2006 15:31, [EMAIL PROTECTED] wrote:
> When a table is indexed, INSERT performance is logorithmic in the
> number of rows in the table and linear in the number of indices.
> This is because entries have to be inserted into the index in
> sorted order (otherwise it
/expected for sqlite, or can this be solved?
--
Dr Peter De Rijk E-mail: [EMAIL PROTECTED]
Bioinformatics Unit Tel. +32 3 265 10 08
Department of Molecular Genetics VIB8Fax. +32 3 265 10 12
University of Antwerphttp
3 matches
Mail list logo