>> The  TtiVisitorUpdate.AfterExecSQL in my installation has no virtual in 
>> it.
>> Does it realted to the version of tiOPF? Mine one is tiOPF 2.
> I only use tiOPF v2 and my revision number sits at r1825.  Maybe you 
> should
> get a repository update?

> The revision number can be confirm by browsing the repository with a web
> browser.
>
>   http://tiopf.svn.sourceforge.net/viewvc/tiopf/tiOPF2/Trunk/
>
> or browse the tiVisitorDB.pas file directly:
> http://tiopf.svn.sourceforge.net/viewvc/tiopf/tiOPF2/Trunk/Core/tiVisitorDB.pas?revision=1692&view=markup

I will look at my tiVisitorDB.pas if I need to update

>> I am using MySQL 5.
>
> I know your database is MySQL, but what database components (persistence
> layer in tiOPF) are you using to access the MySQL database. As far as I 
> can
> see, only two persistence layers in tiOPF can talk to a MySQL database:
> Zeos and SqlDB....  which one do you use?

I have installed Zeos 5.
I wonder if Zeos can exceute INSERT followed by SELECT in one go?


>> and }. I think using the integer as primary key field will be faster when
>> joining a 2 million records tables with other tables. I experienced slow
>
> Evaluate if you really need to join 2 million records.... almost always it
> is possible to limit the data you are working with by defining a better
> WHERE clause.

We have a large table that has 2 millions records.
Yes, the join among this table and other main tables and lookup tables will 
have some
WHERE conditions.

>> Excepcially I am migrating a legacy database (desktop one) to a new MySQL
>> database it took a day for data migration.
>
> I have been using GUID's in our products for over 5 years now. I have not
> noticed any speed issues - we are using Firebird database with SqlDB
> persistence layer.
>
> BTW: My record is a data conversion script that took 4 days of non-stop
> data processing to complete. We had to always schedule it to run over
> weekends. It ran on a 4x CPU SMP system with 8GB RAM, 5 disk RAID system
> and MS SQL-Server database. We didn't use GUIDs as primary keys but rather
> the traditional multiple fields forming a primary key. We just had a crap
> load of data. :-)
>
> I also learned very quickly not do to a 'select * from <table>' 
> statements,
> but rather 'select * from <table> where ...' because there was just too
> many records to return.

I have written a data conversion project to migrate our old desktop database 
to MySQL, running on my desktop pc for exactly 1 day.
That was very slow in that 2 million records table when it passed the first 
1 million records. Might be due to the index page relocating, splitting? 
Because the GUID (primary key) generated value is random, not as sequential 
as auto-incremented? 




------------------------------------------------------------------------------
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
_______________________________________________
tiOPF-talk mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/tiopf-talk

Reply via email to