如果使用表扫描，那么聚集就不会真正有益，因为DB 2仍然必须读取所有的行，那么物理排序不很重要了。

If a table scan is used, the clustering would not really be beneficial because DB2 still has to read all rows and then the physical ordering is not very important.

youdao

如果使用表扫描，那么聚集就不会真正有益，因为DB 2仍然必须读取所有的行，那么物理排序不很重要了。

If a table scan is used, the clustering would not really be beneficial because DB2 still has to read all rows and then the physical ordering is not very important.

youdao