Behavior Changes

When alias name is same as the original column name like “select year(birthday) as birthday” and use it in group by, order by , having clause, doris’s behavior is different from MySQL in the past. In this release, we make it follow MySQL’s behavior. Group by and having clause will use original column at first and order by will use alias first. It maybe a litter confuse here so there is a simple advice here, you’d better not use an alias the same as original column name.

Features

Add support of murmur_hash3_64.

Improvements

Add timezone cache for convert_tz to improve performance.

Sort result by tablename when call show clause. #14492

Bug Fix

ColumnVector::insert_date_column may crashed. #14839

Update high_priority_flush_thread_num_per_store default value to 6 and it will improve the load performance.

Fix quick compaction core. #14731

Partition column is not duplicate key, spark load will throw IndexOutOfBounds error.

Fix create table like when having sequence column. #14511

Using avg rowset to calculate batch size instead of using total_bytes since it costs a lot of cpu.

Fix right outer join core with conjunct. #14821

Optimize policy of tcmalloc gc. #14738