[
https://issues.apache.org/jira/browse/TAJO-838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14183692#comment-14183692
]
ASF GitHub Bot commented on TAJO-838:
-------------------------------------
Github user hyunsik commented on the pull request:
https://github.com/apache/tajo/pull/192#issuecomment-60460566
This work looks great. Please keep going! Actually, this work already has
or will face many challenges. It is definitely worthy to challenge the problems.
I leave some comments. Actually, I just look over the design and the
overall approach because this patch will be committed to ```index_support```
branch. When the patch is merged to ```master``` branch, I'll review it in
implementation level.
In addition, you seem to already have experiment environments. If possible,
could you share the performances with varying selectivity? It's not requirement
for this patch. But, we need to estimate the performance gain when we adopt
indexing approach. Although your benchmark result looks nice,
````0.000045139%``` is too low selectivity in practice.
As we discussed, index size is very important. If we don't have efficient
index structure, this feature won't be used in practice. As we discussed, we
need to investigate more efficient ones.
> Improve query planner to utilize index
> --------------------------------------
>
> Key: TAJO-838
> URL: https://issues.apache.org/jira/browse/TAJO-838
> Project: Tajo
> Issue Type: Sub-task
> Components: planner/optimizer
> Reporter: Jihoon Son
> Assignee: Jihoon Son
> Priority: Minor
>
> Index can improve the query performance when the selectivity of query is high.
> Thus, query planner should decide whether index is used or not for a given
> query.
> The selectivity can be guessed using statistics.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
