You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Commit 09c3e15
Browse filesBrowse the repository at this point in the historyBrowse files
[MT][FEATURE] spark load in MT from 111-rc3 to 115:
change list
1.add broker plus manifest.yaml
2.[MT] spark load for mt
3.[MT] Adapt MT internal spark/yarn commands and configurations
4.[MT] add custom properties for spark load etl & del tmp hive table
5.[MT] delete spark delete spark repository and archive & improve etl job log
6.[MT] feature(sparkload): support bitmap encode features in spark load
7.[MT] feature(sparkload parquet): disable parquet dictionary
8.feature(sparkload): support bitmap binary data from hive in spark load
9.[MT] feature(sparkload): add tolas-output dependency in SparkDpp
10.fix(spark load): resolve args conflict between skip_null_value and map_side_join
refactor(spark load): refactor function name
each commit detail are listed in this branch:
https://dev.sankuai.com/code/repo-detail/data/palo/commit/list?branch=sparkload-14-update-details
or in branh 13:
https://dev.sankuai.com/code/repo-detail/data/palo/commit/list?branch=refs%2Fheads%2F13
Some Spark Load changes in 0.15 to 1.1:
[MT][FIX][SPARKLOAD] fix bug when partition_id exceeds integer range in spark load (apache#9073)
[MT][FIX][SPARKLOAD] fix `getHashValue` of string type is always zero in spark load (apache#9135)
[MT][TMP][SPARKLOAD] support `custom.global.dict.table` in spark load
[MT][FEATURE][SPARKLOAD] support retry-strategy when get the spark elt job state timeout
[MT][SPARKLOAD] hive table name start with tmp key word and its size should be no longer than 128
[MT][TMP][SPARKLOAD] fix min_value will be negative number when `maxGlobalDictValue` exceeds integer range (apache#9436)
detail commmits' content could be found in this branch:
https://dev.sankuai.com/code/repo-detail/data/palo/commit/list?branch=refs%2Fheads%2F14
[MT][TMP][FIX] fix UT in spark load
[MT][FEATURE] feature(spark-dpp version): add version file for spark-dpp
add spark-dpp commit id as version file when build FE
[MT][SPARKLOAD] some fixes from 15 to 1.1 by wangbo36
1 not connect hive metastore when create hive table
2 avoid cast from string to bitmap expr
3 cast bytebuffer to buffer
4 handle exception in ut
0 commit comments