[
https://issues.apache.org/jira/browse/HIVE-28651?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Butao Zhang updated HIVE-28651:
-------------------------------
Description:
Hadoop3.3.5+ （HADOOP-18103）added the Hadoop Vectored IO feature which can be
high-performance against cloud storage.
Apache ORC ORC-1251 and Apache Parquet PARQUET-2171 have integrated this great
feature, so we can try make it work in Apache Hive.
We need to upgrade ORC&Parquet first, then try to test if the Vectored IO can
work.
# HIVE-28650 Upgrade Apache ORC version to 2.0.3
# HIVE-28625 Upgrade Apache Parquet version to 1.14.4
was:
Hadoop3.3.5+ （HADOOP-18103）added the Hadoop Vectored IO feature which can be
high-performance against cloud storage.
Apache ORC ORC-1251 and Apache Parquet PARQUET-2171 have integrated this greate
feature, so we can try make it work in Apache Hive.
We need to upgrade ORC&Parquet first, then try to test if the Vectored IO can
work.
# HIVE-28650 Upgrade Apache ORC version to 2.0.3
# HIVE-28625 Upgrade Apache Parquet version to 1.14.4
> Make Hadoop Vectored IO work in Apache Hive
> -------------------------------------------
>
> Key: HIVE-28651
> URL: https://issues.apache.org/jira/browse/HIVE-28651
> Project: Hive
> Issue Type: Improvement
> Security Level: Public(Viewable by anyone)
> Reporter: Butao Zhang
> Priority: Major
>
> Hadoop3.3.5+ （HADOOP-18103）added the Hadoop Vectored IO feature which can be
> high-performance against cloud storage.
> Apache ORC ORC-1251 and Apache Parquet PARQUET-2171 have integrated this
> great feature, so we can try make it work in Apache Hive.
> We need to upgrade ORC&Parquet first, then try to test if the Vectored IO can
> work.
> # HIVE-28650 Upgrade Apache ORC version to 2.0.3
> # HIVE-28625 Upgrade Apache Parquet version to 1.14.4
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
