percona-discussion team mailing list archive
-
percona-discussion team
-
Mailing list archive
-
Message #00557
Re: simple optimizations
On Thu, Apr 30, 2009 at 8:59 PM, Baron Schwartz <baron@xxxxxxxxx> wrote:
>
> Yeah, but... changing checksum would instantly result in all pages
> being "corrupt", no?
Yes, so it warrants a new Innodb file format: antelope, barracuda, .... checksum
As soon as I get a nehalem I will find out how much faster HW support
for crc helps.
Eventually, xtra-backup may get far enough along that you need to
introduce a new InnoDB file format and at that point you can change
the file format.
>
>> But there's an additional issue with the whole checksum foo. The checksum is
>> right next to the data, and while CRCs are decent at detecting corruption,
>> the proximity can also hide some occurrances. ZFS has the nifty trick of
>> storing checksums separately, and I think that's quite brilliant. Not saying
>> it's easy to do for InnoDB or even if it's desirable - just want to put it
>> out there so we can discuss.
>
> I think it's a really good design. It seems "hard" at this stage in
> InnoDB's life, though.
>
> What about PBXT or Maria? I don't even know if they checksum data.
>
--
Mark Callaghan
mdcallag@xxxxxxxxx
References