> -Original Message-
> From: Harald Fuchs [mailto:[EMAIL PROTECTED]
> Sent: 30 September 2004 12:16
> To: [EMAIL PROTECTED]
> Subject: Re: Indexing problem with UTF8 in 4.1.4?
>
> In article
> <[EMAIL PROTECTED]>,
> Kevin Cowley <[EMAIL PROTECTED]>
In article <[EMAIL PROTECTED]>,
Kevin Cowley <[EMAIL PROTECTED]> writes:
> If I knew why I wouldn't be asking. Now by our reconing the key of the
> fields is 343 bytes, encoding in UTF8 makes that key 343 bytes Not 1000
> since under utf8 each character is encode in 8 bits.
What makes you think s
> Now by our reconing the key of the
> fields is 343 bytes, encoding in UTF8 makes that key 343 bytes Not
> 1000
> since under utf8 each character is encode in 8 bits. If it is
> converting to
> utf16 internally then the key would be 328*2+5+4 which is not >1000
> bytes.
>
If you only use the as
Kevin Cowley
R&D
Tel: 0118 902 9099 (direct line)
Email: [EMAIL PROTECTED]
Web: http://www.alchemetrics.co.uk
> -Original Message-
> From: gerald_clark [mailto:[EMAIL PROTECTED]
> Sent: 29 September 2004 17:29
> To: Kevin Cowley
> Cc: [EMAIL PROTECTED]
> Subject
Kevin Cowley wrote:
Running 4.1.4 with a database that has a default encoding of UTF8
If we execute the following we get an error.
CREATE TABLE idxbe_resident (
urn INT UNSIGNED NOT NULL,
keyAddress_Part1 CHAR(5) BINARY NOT NULL,
dataPerson_Name CHAR(60),
dataAddress_Part1 CHAR(140),
dataAddress_P
Running 4.1.4 with a database that has a default encoding of UTF8
If we execute the following we get an error.
CREATE TABLE idxbe_resident (
urn INT UNSIGNED NOT NULL,
keyAddress_Part1 CHAR(5) BINARY NOT NULL,
dataPerson_Name CHAR(60),
dataAddress_Part1 CHAR(140),
dataAddress_Part2 CHAR(128),
IND
Hi,
We have given up trying to understand why the problem (please see below)
occurs, and have moved on to see what will be the best workaround. We have
three different suggestions. Alternatives 1 and 2 below are under the
assumption that setting the max_rows parameter to 10.000.000.000 caused the
Hello!
I hope someone can help us with a problem we have:
We had a table that was 35G big and the index file was about 16G (ca 120
million records). We ran into a problem where the database produced an "no
more room in index file" error.
To fix this we regenerated the table and set max_rows t
umn I use
is VARCHAR(40), and I am not sure when I use
create index idx1 on transaction(vendor_id(20));
will help???
-Original Message-
From: Sergei Golubchik [mailto:[EMAIL PROTECTED]]
Sent: Friday, October 26, 2001 10:53 AM
To: Wai Lee
Cc: [EMAIL PROTECTED]
Subject: Re: Indexing Prob
Hi!
On Oct 26, Kyle Hayes wrote:
> On Friday 26 October 2001 07:18, Wai Lee wrote:
>
> > I dig through the manual and changed any possible settings
> >
> > set-variable= max_heap_table_size=2000M
> > set-variable= key_buffer=2500M
> > set-variable= max_allowed_packet=1M
> > set-varia
From: "Wai Lee" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, October 26, 2001 4:18 PM
Subject: Indexing Problem
> Hi all,
>
> Can someone tell me how to speed up the index creation???
>
> I am trying to build an index for a 13,875,354 records
On Friday 26 October 2001 07:18, Wai Lee wrote:
> I dig through the manual and changed any possible settings
>
> set-variable= max_heap_table_size=2000M
> set-variable= key_buffer=2500M
> set-variable= max_allowed_packet=1M
> set-variable= table_cache=512
> set-variable= sort_
Hi!
On Oct 26, Wai Lee wrote:
> Hi all,
>
> Can someone tell me how to speed up the index creation???
>
> I am trying to build an index for a 13,875,354 records(13 million) table
> with 176,322 distinct vendor_id(VARCHAR(40)) in the table.
>
> the existing size of the transaction table:
> tran
Hi all,
Can someone tell me how to speed up the index creation???
I am trying to build an index for a 13,875,354 records(13 million) table
with 176,322 distinct vendor_id(VARCHAR(40)) in the table.
the existing size of the transaction table:
transaction.MYD = 2,128,954,624 bytes
transaction.MYI
With the correct indexes this query should run in less than a second.
Basically what you should do is make an index for every column that is used
in the where clause. For instance, if you had a query ... WHERE A=X AND B=Y
AND C=Z (A,B,and C are in the same table) you would create an index (A,B,C)
Folks,
I am with Security Managment Consulting, and we have a liscensed copy of
mysql. I don't have the lisc. # on hand.
We have a very large database in mysql, and one of the tables (call it
badge_history) has about 4,000,000 records, with the potiental to get
much bigger...
Now, we are using
16 matches
Mail list logo