Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Plain View
Hive, mail # dev - Question about documentation for Hive in general


+
Lars Francke 2012-06-12, 23:27
+
Edward Capriolo 2012-06-13, 02:10
+
Lars Francke 2012-06-14, 12:04
Copy link to this message
-
Re: Question about documentation for Hive in general
Edward Capriolo 2012-06-14, 16:12
The indexes would not make sense on storage handler tables. They
attempt to make an index of data->dfs block. Then that information is
used to prune blocks from the job, this would not make sense for
storage handlers because the data is not directly backed by blocks.

Edward

On Thu, Jun 14, 2012 at 8:04 AM, Lars Francke <[EMAIL PROTECTED]> wrote:
> Edward,
>
> thanks for your reply.
>
>> As you may have heard myself and some others (Dean, Jason) are working
>> on the programming hive book.
>
> Yeah, heard about that and I'm looking forward to it!
>
>> I recently stumbed onto the really amazing thrift raw bytes serde. Who
>> knew that hive table can be defined only with the named of a Thrift
>> class and magically the column meta data just auto-populates?
>
> It's exactly that kind of stuff that I mean. There are so many great
> things in Hive that no one knows about :)
>
>> 1) They work and can accelerate some types of queries but there is
>> more work to be done
>
> Okay, thanks. Any idea if they work with external (e.g. HBase) tables?
> I found conflicting information.
>
>> 2) I dont know,
>
> I'll ask again maybe on the user list
>
>> 3) No the docs on binary are not fully accurate
>
> Same as 2) then but I'll look at the patch first.
>
> I might come back to the list or JIRA with questions like this in the
> future. I hope that's okay.
>
> cheers,
> Lars