I often get asked questions about this topic, so I've put together a wiki page
which expresses some of my thoughts on it:
https://cwiki.apache.org/confluence/display/Hive/BecomingACommitter
Let me know if there are points you'd like to add, or where you see it
differently.
JVS
Thanks! Can Hive index LZO compressed files then? LZO compression isn't
part of Cloudera's release, right?
On 03/08/2011 19:38, yongqiang he wrote:
unfortunately it does not, because can not split .gz file.
2011/8/3 Martin Konicek:
Hi,
can indexes work on gzipped files?
The index gets build
Hello,
Is this not the forum for this type of question? Is there another
forum someone recommends?
Thanks,
Ajo.
On Tue, Aug 2, 2011 at 9:35 AM, Ajo Fod wrote:
> Hello Hive Gurus,
>
> I am not sure if my system is using the sorting feature.
>
> In summary:
> - I expected to save time on the sort
Hi all,
I have used compact index for my table and the response time is
same for a query with as well as without index now. Previously, it was
showing improvement. I just changed some parameters to increase heap size
and then it is behaving weird. so, how can I make sure that my query is
u
unfortunately it does not, because can not split .gz file.
2011/8/3 Martin Konicek :
> Hi,
>
> can indexes work on gzipped files?
>
> The index gets build without errors using
> ALTER INDEX syslog_index ON syslog PARTITION(dt='2011-08-03') REBUILD;
>
> but when querying, no results are returned (a
Hi Carl,
Thank you very much!
Cheers, Neerja
On Tue, Aug 2, 2011 at 11:31 PM, Carl Steinbach wrote:
> Hi Neerja,
>
> Ordinarily a POM for the hive-hbase-handler would be available
> from Cloudera's Maven repository, but it looks like we haven't
> generated one. We will get this resolved in CDH
Hi,
can indexes work on gzipped files?
The index gets build without errors using
ALTER INDEX syslog_index ON syslog PARTITION(dt='2011-08-03') REBUILD;
but when querying, no results are returned (and no errors reported). The
query should be correct because with plaintext files it works.
Bes
Hi Edward,
I think option 1 with hive.rc is the path we want to take.
I am really not looking for web-client users ton install UDF jars. Our use-case
is:
We as the Infrastructure team have written a UDF that we want our internal
web-client customers to use without them having to install any jars.
On Wed, Aug 3, 2011 at 2:55 AM, Ayon Sinha wrote:
> So if I have a web-client calling the Hive server, will step 1 & 2 need to
> be executed for every call as a sequence (that would be inconvenient). I
> really want this UDF to be available to every user of our web interface.
> They shouldn't nee
