On Tue, 2006-06-27 at 16:14 +0400, Andrew Sagulin wrote:
> Result showed up that there were no page seq scan at all - true random access
> only.
> The simple model which can explain the situation: the sequence of numbers 2,
> 1,
> 4, 3, 6, 5, ..., 100, 99 has correlation about 0,9994. Let's imagi
> I think the problem is probably that you're sorting two dozen CHAR
> columns, and that in many of the rows all these entries are '' forcing
> the sort code to compare all two dozen columns (not so)?
Yes, most of columns return empty strings.
I changed empty strings to null, casted to varchar an
unregister
-- Best,Gourish Singbal
Hello all
I have a big amount of phone calls data (280M records, 25 Gbytes).The best
decision
for this task is partitioning and I use it now. But at first I tried put all
data in a single table indexed by call date&time. Because of nature of the
data the records clustered by date and near ordered
On Mon, 2006-06-26 at 17:20 -0400, Michael Stone wrote:
> On Mon, Jun 26, 2006 at 08:33:34PM +0100, Simon Riggs wrote:
> >of the SQL standard, so being unaware of them when using SQL is strange
> >to me.
>
> Welcome to the world of programs designed for mysql. You'll almost never
> see them batc
