maria-developers team mailing list archive

[Stewart Smith <stewart@xxxxxxx>]

On Wed, Jul 08, 2009 at 03:53:15AM -0400, Peter Benjamin Volk wrote:
> Stewart Smith wrote:
> > The idea being to move the handler to be a cursor on a table, with
> > actions not pertaining that to reside in StorageEngine (e.g. DDL).
> >   
> That sounds like an interesting Idea. This would though involve quite a
> few changes. Would the curser then still be row based? Or might you
> convert to a more column oriented way?

We had a more column oriented API in NDB, and found that going to
NdbRecord (very much row based) got us a pretty big performance boost.

Or are you thinking for column based engines?

> > There's also the (now rather old) change to drop table return code.
> >
> > The next thing that will move into the StorageEngine is metadata
> > handling with the engine being able to be responsible for its own
> > (table) metadata.
> >   
> That would be wonderful. Does the Meta data include statistics on the
> table (e.g. selectivity estimates etc) that would be needed for the
> optimizers cost model?

Not yet.

I have been thinking in the back of my mind how to make this better though.

> > So, is there somebody interested in working with me to have the
> > MySQL/MariaDB API evolve in the same way?
> >   
> Is there a targeted MySQL version for this?

Not at all up to me.

I'm doing things in Drizzle.... it's really up to the MySQL folks if
they want to pull in the improvements too.

-- 
Stewart Smith