-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/28964/#review67616
-----------------------------------------------------------


Nice work Sergio!!

I know that it doesn't fit perfectly into the JMH model but I think we have to 
write a non-trival amount of records such as 1000 rows in order to get much 
benefit. Can we try that?


itests/hive-jmh/pom.xml
<https://reviews.apache.org/r/28964/#comment111684>

    It looks like in this file 1 tab = 4 spaces whereas in Hive I think we 
typically say 1 tab = 2 spaces



itests/hive-jmh/src/main/java/org/apache/hive/benchmark/storage/ColumnarStorageBench.java
<https://reviews.apache.org/r/28964/#comment111683>

    During class initialization let's create an array of 100 random values for 
each type and then we can iterate through that array for each call to this 
method.
    
    Otherwise columnar formals will lead to unrealistic comppression for 
storing the same values over and over. For example both parquet and orc should 
be able to collapse a column consist of the integer 1 to a trivial amount of 
data.


- Brock Noland


On Jan. 9, 2015, 6:38 p.m., Sergio Pena wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/28964/
> -----------------------------------------------------------
> 
> (Updated Jan. 9, 2015, 6:38 p.m.)
> 
> 
> Review request for hive, Brock Noland and cheng xu.
> 
> 
> Bugs: HIVE-8121
>     https://issues.apache.org/jira/browse/HIVE-8121
> 
> 
> Repository: hive-git
> 
> 
> Description
> -------
> 
> This is a new tool used to test ORC & PARQUET file format performance.
> 
> 
> Diffs
> -----
> 
>   itests/hive-jmh/pom.xml PRE-CREATION 
>   
> itests/hive-jmh/src/main/java/org/apache/hive/benchmark/storage/ColumnarStorageBench.java
>  PRE-CREATION 
>   itests/pom.xml 0a154d6eb8c119e4e6419777c28b59b9d2108ba0 
> 
> Diff: https://reviews.apache.org/r/28964/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Sergio Pena
> 
>

Reply via email to