Hi Alberto,
Thanks for your comments! In many cases the data is imported to Hadoop in
T+1 mode. Especially when everyday's data is tens of GB, it is
reasonable to partition the Hive table by date. The problem is whether it
worth to keep a long history data in Hive; Usually user only keep a couple
Yerui,
You're correct, serialize the dictionary isn't a good idea; I will try to
initialize these big objects inner executors, instead of transfering them
from driver; I will get back to you if have problem. Thanks!
2017-01-25 15:56 GMT+08:00 Yerui Sun :
> Hi，shaofeng,
>Sorry for my slow res
Make the price column a dimension and you should be able to do the query.
However double being a very high cardinality type may not work well (high
expansion rate, long build time etc). If you could lower the price's
cardinality by rounding to integer or even tens/hundreds, the resulted cube
will b
It's in Spanish, but the picture is very clear
https://www.linkedin.com/pulse/qlik-cloudera-bigdatasmartdataanalytics-felipe-trigo?trk=hp-feed-article-title-publish
Be careful about partition by "FLIGHTDATE"
>From https://github.com/albertoRamon/Kylin/tree/master/KylinPerformance
*"Option 1: Use id_date as partition column on Hive table. This have a big
problem: the Hive metastore is meant for few hundred of partitions not
thousand (Hive 9452 there is an ide
The solution of Li Yand works works from CDH 5.4, but If your production
env is HBase 1.2, dreaming is free ;) , you can also try:
https://issues.apache.org/jira/browse/HBASE-10070
2017-01-25 5:52 GMT+01:00 Li Yang :
> Try google 'hbase read replica'.
>
> Cheers
>
> On Tue, Jan 17, 2017 at 1
Hello,
A new document is added for the practices of cube build. Any suggestion or
comment is welcomed. We can update the doc later with feedbacks;
Here is the link:
https://kylin.apache.org/docs16/howto/howto_optimize_build.html
--
Best regards,
Shaofeng Shi 史少锋
+1, nice work.
2017-01-25 15:59 GMT+08:00 Yerui Sun :
> +1
>
> Use hbase 1.x as default version really make sense.
>
> However, we also should support 0.98 version as past, since upgrading is
> not easy in some production environments.
>
> > 在 2017年1月23日，11:41，nichunen 写道：
> >
> > +1
> >
> > Hba
XIE FAN created KYLIN-2417:
--
Summary: Compare the performance between HDFSMetaStore and
HBaseMetaStore
Key: KYLIN-2417
URL: https://issues.apache.org/jira/browse/KYLIN-2417
Project: Kylin
Issue Typ
+1
Use hbase 1.x as default version really make sense.
However, we also should support 0.98 version as past, since upgrading is not
easy in some production environments.
> 在 2017年1月23日，11:41，nichunen 写道：
>
> +1
>
> Hbase 1.x is more commonly used in our clients.
>
>
> George/倪春恩
>
> Mob
10 matches
Mail list logo
