Comments inline ...

On Tue, Mar 5, 2013 at 9:15 AM, ddorian <[email protected]> wrote:

>
>    - why do people want to discover large rows?
>
> Hypertable (and other Bigtable based DBs) have a limitation on the maximum
size of a row.  A row cannot grow larger than a single range.  The default
range split size is 256MB.  Once a row has hit this limit, it will stop
accepting updates and send back a ROW_OVERFLOW error.  Some applications
want to anticipate this situation before it happens and handle large rows
differently.

>
>    - if this will have overhead make it optional ?
>
> It will have no overhead.  Pre-existing applications (that don't use the
feature) will see no impact.

>
>    - other ways that pseudo-tables will be used?
>
> I can imagine using the pseudo-table interface for exposing CellStore
trailer data as well as in-memory statistics from AccessGroups, Ranges, and
RangeServers.

>
>    - very happy to see the first signs of advanced queries
>
> Me too.  The 0.9.8 release and beyond will mostly be focused on advanced
query capabilities.  We'll get a roadmap document in place soon.

- Doug

-- 
You received this message because you are subscribed to the Google Groups 
"Hypertable Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/hypertable-dev?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to