Usually, tables with shorter rows yield better performance than those with longer rows because disk I/O is performed in pages, not in rows. The shorter the rows of a table, the more rows occur on a page. The more rows per page, the fewer I/O operations it takes to read the table sequentially, and the more likely it is that a nonsequential access can be performed from a buffer.
The entity-relationship data model puts all the attributes of one entity into a single table for that entity. For some entities, this strategy can produce rows of awkward lengths. To shorten the rows, you can break columns into separate tables that are associated by duplicate key values in each table. As the rows get shorter, query performance should improve.
Enterprise Edition Home | Express Edition Home | [ Top of Page | Previous Page | Next Page | Contents | Index ]