Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Plain View
Hive >> mail # dev >> Review Request: New code for VectorizedRowBatch to form basis of vectorized query execution


+
Eric Hanson 2013-04-17, 20:50
+
Carl Steinbach 2013-04-17, 21:47
Copy link to this message
-
Re: Review Request: New code for VectorizedRowBatch to form basis of vectorized query execution


> On April 17, 2013, 9:47 p.m., Carl Steinbach wrote:
> > ql/src/java/org/apache/hadoop/hive/ql/exec/vector/DoubleColumnVector.java, line 33
> > <https://reviews.apache.org/r/10592/diff/1/?file=282042#file282042line33>
> >
> >     Is there a reason these variables are public?

The major fields of the column vector and vectorized row batch types are public for fast access in the inner loop of query execution. This is by design. This is described in code comments.
> On April 17, 2013, 9:47 p.m., Carl Steinbach wrote:
> > ql/src/java/org/apache/hadoop/hive/ql/exec/vector/LongColumnVector.java, line 33
> > <https://reviews.apache.org/r/10592/diff/1/?file=282043#file282043line33>
> >
> >     Public?

see previous comment re: public fields
> On April 17, 2013, 9:47 p.m., Carl Steinbach wrote:
> > ql/src/java/org/apache/hadoop/hive/ql/exec/vector/VectorizedRowBatch.java, line 36
> > <https://reviews.apache.org/r/10592/diff/1/?file=282044#file282044line36>
> >
> >     Public?

see previous comment re: public fields
> On April 17, 2013, 9:47 p.m., Carl Steinbach wrote:
> > ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ColumnVector.java, line 30
> > <https://reviews.apache.org/r/10592/diff/1/?file=282041#file282041line30>
> >
> >     Should these methods be abstract? If not, then I think you want to throw an UnsupportedOperationException("Not Implemented") instead of doing an assert or returning a string.

methods deleted
- Eric
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/10592/#review19350
-----------------------------------------------------------
On April 18, 2013, 1:27 a.m., Eric Hanson wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/10592/
> -----------------------------------------------------------
>
> (Updated April 18, 2013, 1:27 a.m.)
>
>
> Review request for hive.
>
>
> Description
> -------
>
> New code for VectorizedRowBatch to form basis of vectorized query execution
>
>
> This addresses bug HIVE-4284.
>     https://issues.apache.org/jira/browse/HIVE-4284
>
>
> Diffs
> -----
>
>   ql/src/java/org/apache/hadoop/hive/ql/exec/vector/BytesColumnVector.java PRE-CREATION
>   ql/src/java/org/apache/hadoop/hive/ql/exec/vector/ColumnVector.java PRE-CREATION
>   ql/src/java/org/apache/hadoop/hive/ql/exec/vector/DoubleColumnVector.java PRE-CREATION
>   ql/src/java/org/apache/hadoop/hive/ql/exec/vector/LongColumnVector.java PRE-CREATION
>   ql/src/java/org/apache/hadoop/hive/ql/exec/vector/VectorizedRowBatch.java PRE-CREATION
>   ql/src/test/org/apache/hadoop/hive/ql/exec/vector/TestVectorizedRowBatch.java PRE-CREATION
>
> Diff: https://reviews.apache.org/r/10592/diff/
>
>
> Testing
> -------
>
>
> Thanks,
>
> Eric Hanson
>
>

+
Eric Hanson 2013-04-18, 01:27
+
Eric Hanson 2013-04-18, 19:41
+
Carl Steinbach 2013-04-18, 08:50