Greg, Thank you very much for your help.
> An index the full size of the distinct key is probably very quick especially > if it only has 20 distinct values. OTOH, if you already know you have a > small number of distinct values, could you just store them normalized in a > different table? Actually, it is. It has some tables: features ===== - id - description groups ===== - id - description products ====== - id - description - group_id - feature_id And Iīll use a SELECT like this: SELECT DISTINCT features.description FROM products LEFT JOIN features ON (products.feature_id = features.id) WHERE products.group_id = $var_group AND products.features_id > 0 The table products should have a million of records, but the filtered query should goes over a thousand records (filtered by group_id) and return about 20 distinct lines. Itīs my situation but I donīt know how heavy such query is for the database and how viable such query is. Thanks in advance, Ronan -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
