You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@drill.apache.org by "Jerry Ylilammi (JIRA)" <ji...@apache.org> on 2015/12/14 10:20:46 UTC

[jira] [Created] (DRILL-4193) All Parquet columns are loaded when querying S3

Jerry Ylilammi created DRILL-4193:
-------------------------------------

             Summary: All Parquet columns are loaded when querying S3
                 Key: DRILL-4193
                 URL: https://issues.apache.org/jira/browse/DRILL-4193
             Project: Apache Drill
          Issue Type: Bug
          Components: Storage - Parquet
    Affects Versions: 1.3.0
            Reporter: Jerry Ylilammi


Drill starts downloading all data from S3 not making use of Parquet being columnar format and only loading required column.

Query:
{code}SELECT DISTINCT data.measurement.cid 
FROM mys3bucket.`test/datatable` AS data;{code}

Parquet file:
{code}...
measurement:
.cid:                  INT64 GZIP DO:0 FPO:9380624 SZ:110/76/0.69 VC:32327 ENC:BIT_PACKED,PLAIN_DICTIONARY,RLE
...{code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)