[
https://issues.apache.org/jira/browse/DRILL-7706?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17085973#comment-17085973
]
ASF GitHub Bot commented on DRILL-7706:
---------------------------------------
paul-rogers commented on issue #2060: DRILL-7706: Implement Drill RDBMS
Metastore
URL: https://github.com/apache/drill/pull/2060#issuecomment-615394559
@arina-ielchiieva, this is great! I wonder, will this PR handle a particular
use case that I'm seeing?
The metastore works for files. We are seeing more cases where users want to
read other data sources: HTTP, Mongo, whatever. These sources deliver JSON,
which can be ambiguous. We want to use a provided schema to resolve
ambiguities. (That work is ongoing in updating the JSON reader, etc.) So, we
need a place to store the provided schema.
Recall that, for files, we put the provided schema in the same folder as the
data. For the HTTP plugin, say, there is no directory to place the files.
So, can we use the DB-backed metastore? Will there be a standard way for a
plugin to map its concept of a table to an entry in the metastore which can
hold the provided schema?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Drill RDBMS Metastore
> ---------------------
>
> Key: DRILL-7706
> URL: https://issues.apache.org/jira/browse/DRILL-7706
> Project: Apache Drill
> Issue Type: New Feature
> Affects Versions: 1.17.0
> Reporter: Arina Ielchiieva
> Assignee: Arina Ielchiieva
> Priority: Major
> Fix For: 1.18.0
>
>
> Currently Drill has only one Metastore implementation based on Iceberg
> tables. Iceberg tables are file based storage that supports concurrent writes
> / reads but required to be placed on distributed file system.
> This Jira aims to implement Drill RDBMS Metastore which will store Drill
> Metastore metadata in the database of the user's choice. Currently,
> PostgreSQL and MySQL databases are supported, others might work as well but
> no testing was done. Also out of box for demonstration / testing purposes
> Drill will setup SQLite file based embedded database but this is only
> applicable for Drill in embedded mode.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
