On Sun, Jul 18, 2010 at 04:31:20PM +0100, Simon Slavin scratched on the wall:
> On 18 Jul 2010, at 2:56pm, Peng Yu wrote:
> > So, in general, what the index I should use depends on what the select
> > statement I might use?
> Exactly. There is one index which is ideal for each SELECT statement.
2010/7/18 Peng Yu :
>>> This question may be trivial. But I'm wondering, whether I should
>>> create one index on multiple columns or create multiple indexes, each
>>> of which is on a single column.
>
> select * from test group by value1,value2;
> select * from test group by value1||value2;
>
> Fo
On 18 Jul 2010, at 2:56pm, Peng Yu wrote:
> On Sun, Jul 18, 2010 at 12:08 AM, Simon Slavin wrote:
>>
>> On 18 Jul 2010, at 5:46am, Peng Yu wrote:
>>
>>> I'm wondering, whether I should
>>> create one index on multiple columns or create multiple indexes, each
>>> of which is on a single column.
On Sun, Jul 18, 2010 at 12:08 AM, Simon Slavin wrote:
>
> On 18 Jul 2010, at 5:46am, Peng Yu wrote:
>
>> This question may be trivial. But I'm wondering, whether I should
>> create one index on multiple columns or create multiple indexes, each
>> of which is on a single column.
>>
>> The condition
On 18 Jul 2010, at 5:46am, Peng Yu wrote:
> This question may be trivial. But I'm wondering, whether I should
> create one index on multiple columns or create multiple indexes, each
> of which is on a single column.
>
> The condition of the 'select' statement can be a logic operation on
> any of
This question may be trivial. But I'm wondering, whether I should
create one index on multiple columns or create multiple indexes, each
of which is on a single column.
The condition of the 'select' statement can be a logic operation on
any of the column of the table and their combinations. I think
6 matches
Mail list logo