QUESTION - What is the real/full proposed grammar for this as it relates
to the containing AsterixDB SQL++ grammar?
I'm asking because this APE (APE 15) seems to only allow (at least as
described) copying a full collection to or from HDFS, where as the S3
case (APE 4) is much richer - it allows copying a query result to HDFS
and also provides a way to structure the results (e.g., into a
date-organized hierarchy). I was expecting this new APE for HDFS to
support the same capabilities, including a SQL++ query as the source and
result structuring, just for a different target external filesystem.
What's the plan?
Thanks,
Mike
On 11/7/24 3:16 PM, Murtadha Hubail wrote:
+1 to implementing this enhancement.
-Murtadha
________________________________
From: Savyasach Reddy Enukonda<esavyared...@gmail.com>
Sent: Friday, November 8, 2024 12:08:51 AM
To:dev@asterixdb.apache.org <dev@asterixdb.apache.org>
Subject: [DISCUSS] Support writing query result to HDFS
Hi all,
I’m reaching out to start a discussion to support writing query results to HDFS.
AsterixDB’s current external datasets implementation is restricted to reading
from unsecured HDFS. We propose enabling the reading and copying of results to
secured-HDFS clusters, similar to the copy and read operations from S3.
APE:https://cwiki.apache.org/confluence/x/I42MEw
Best,
Savyasach Reddy
