phoenix-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From alex kamil <alex.ka...@gmail.com>
Subject Re: Has anyone tried Phoenix-HBase with Pentaho Mondrian?
Date Sat, 31 May 2014 16:12:55 GMT
Feng, I believe this is work in progress, see these threads:
Phoenix with Mondrian
<https://groups.google.com/forum/#!topic/phoenix-hbase-user/-YUde_egPzc>
Phoenix with Tableau
<https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CCkQFjAA&url=http%3A%2F%2Fmail-archives.apache.org%2Fmod_mbox%2Fphoenix-user%2F201404.mbox%2F%253CCAO%3D2E%3Du89oZULiWaH7%3DSxVD%2B7DkgUyAN_6OnK_s6bO%2BMbgu%2BaA%40mail.gmail.com%253E&ei=Jv-JU4_EC8HlsATgn4GYBw&usg=AFQjCNErT89I0c0W_3EQMMp7yuI-fXFmGw&sig2=1nUpLfPw2FqcwYhxd51LQw>

Alex



On Tue, May 27, 2014 at 5:55 AM, Feng Wang <wfeng198209@gmail.com> wrote:

> can you do the query with phoenix mondrian? now I found the same problem.
> mondrian 3.6.1
>
> 在 2013年3月10日星期日UTC+8上午12时03分59秒,Karthikeyan Muthukumarasamy写道:
>>
>> Hi,
>> Has anyone tried Phoenix with Pentaho Mondrian?
>> In our product, we use the community edition of Pentaho Mondrian as an
>> embedded OLAP engine and have our own OLAP framework built over this.
>> Our core data processing is all in Hadoop/HBase and just before the OLAP
>> part, we are now importing all the marts from HBase into an RDBMS (now
>> Oracle), just because Mondrian cannot take data directly from Hadoop/HBase.
>> I have tried the Hive interface in the past but the latency is unbearable!
>> PS: I understand that phoenix today doesnt support Joins and hence would
>> use the degenerate dimension feature in Mondrian to have a flattened out
>> fact table with contains dimensions in itself. Actually, in HBase thats how
>> we would maintain it anyway!
>> Phoenix itself looks fine as a JDBC layer but when I integrate phoenix
>> driver with Mondrian, Phoenix is not able to understand the SQL that
>> Mondrian generates.
>> When I looked into the SQL query its mainly use of apostrophe (') before
>> and after attributes and a mention of table name as itself (like in "select
>> 'sales'.'month' as 'c0' from 'sales' as 'sales' group by 'sales'.'month'
>> order by CASE WHEN 'sales'.'month' IS NULL THEN 1 ELSE 0 END,
>> 'sales'.'month' ASC") which is not acceptable to Mondrian. When I take the
>> query, modify it by removing apostrophe and the use of "as sales" and
>> execute it in Squirrel SQL client, it works fine in Phoenix.
>>
>> Now my question is:
>> I understand that I can built a dialect layer for Phoenix in Mondrian (if
>> its not available!!!) and make these small modifications to the SQL query
>> generated by Mondrian. If I do that, does this mean by using degenerate
>> dimensions, we can make Mondrian work with Phoenix? or am I missing
>> something here?
>>
>> Thanks & Regards
>> MK
>>
>>
>  --
> You received this message because you are subscribed to the Google Groups
> "Phoenix HBase User" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to phoenix-hbase-user+unsubscribe@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
>

Mime
View raw message