Thank you for your return Emmanuel,
Never tested but had lot's of good returns about Maatkit : (
http://www.maatkit.org/ )
Take a look at mk-parallel-dump and mk-parallel-restore
On multi-core architecture you can gain *Nbcore against mysqldump/mysql
(use 1 core do proceed is restore)
A thinks could be to do some sed "transformation" (I don't know java) on
the sql file dumped by mysqldump to add the "atler table" statement just
after the "create table" statement on the corresponding table, so the
script copy an empty table before filling it.
Hope this may help.
Cheers,
--
Damien
>
> Hi Damien,
>
> We do not have control on the order of operations since we just invoke
> mysqldump to generate the dump and replay it as is.
> If you find a more efficient backup tool or other mysqldump options that
> would generate a dump that would be more efficient to replay, don't
> hesitate to let us know.
>
> Thanks for your interest in Sequoia,
> Emmanuel
>
>> I am currently, restoring a 2.8G file dump of a MySQl backend on a
>> 2.10.10
>> sequoia platform.
>> The process is very long at the end of the restoration because of the
>> setting of the auto_increment variable on each tables (I have 3 big
>> tables).
>>
>> the "alter table" statement result in a copy of the whole table in a
>> temporary table ('copy to tmp table' in the state column of the
>> processlist) ... so with a big table it could be very long.
>>
>> I read this bug
>> http://www.mail-archive.com/[email protected]/msg01077.html
>> about this problem.
>>
>> Is it possible for you just to insert the "alter table" after each
>> "create
>> table" instead of after the "insert" statements at the end of the dump
>> script ?
>>
>> Cheers,
>>
>>
>
>
> --
> Emmanuel Cecchet
> FTO @ Frog Thinker
> Open Source Development & Consulting
> --
> Web: http://www.frogthinker.org
> email: [email protected]
> Skype: emmanuel_cecchet
>
> _______________________________________________
> Sequoia mailing list
> [email protected]
> http://forge.continuent.org/mailman/listinfo/sequoia
>
>
_______________________________________________
Sequoia mailing list
[email protected]
http://forge.continuent.org/mailman/listinfo/sequoia
