在Hive的存储桶表中增量加载数据？

Question

I am still learning hive. I have referred few books for understanding concept of Buckets in hive. What i learnt is , if we enforce bucketing , it will create exactly same number of files as number of buckets.

In my Case, i will load the data incrementally in bucketed table five times a day. For ex: if i have table with 16 buckets, then each load it will create 16 files based on hash/samples. So totally for 5 runs , 80 files will created.

My Question is , if i have table with 16 buckets defined on it with 80 files 
in HDFS, will it going to give bucketing benefits ?

Answer 1

do you create a different table for each incremental loading?

which Hadoop distribution are you using?

I'm using that same strategy and every incremental loading generates (and override) the same number of buckets that i defined.

When we have permissions problems have duplicated files, because hive tables were created with hive user and populating was made with another (hdfs).

Look at in your /user/hive/warehouse directory for the owner/permissions of table directory , afterward look at in the subdirectories for the same (owner/permissions)