[TYPO3-dev] TYPO3 4.1 INNODB

[R. van Twisk]

Steffen Müller wrote:
> Hi Dmitry,
>
> thank you for your explanation.
>
> On 01.03.2007 11:01 Dmitry Dulepov wrote:
>   
>> But there is one more advantage for Innodb. Myisam keeps indexes
>> separately from row data. Thus if row pointer is found, myisam engine
>> has to do one  more seek to data. Innodb keeps indexes together with
>> rows, so it does not have to seek again.
>>
>>     
>
>   
Normally this is a disadvantage, (But I don't know innodb very well)
I have been working with a DB system on which I could store the indexes
on a separate raid array. This greatly speeded up the DB.

I can imagine (what steffen said) is true for the primary key.. I am not 
sure
why they would do it like that apart from that one less seek.

(I think that was OT....)
Ries


> This is only true for the primary index. Secondary indices are not
> clustered and thus need also one more seek, too. (see Zawodny/Balling)
> But I don't have a clue, if this is relevant for TYPO3.
>
>