Hive group by 数据倾斜问题处理

insert overwrite table raw_search_behavior partition(dt='2022-10-19')
select cookie_id,event_query,count(*) as cnt,
max(from_unixtime(unix_timestamp(event_time), 'yyyyMMddHHmmss')) as last_dt
from raw_query
where dt >= '20220718' and dt <= '20221019' and event_query is not null
group by cookie_id,event_query

insert overwrite table raw_search_behavior partition(dt='2022-10-19')
select split(tkey,'_')[1] as cookie_id,event_query, sum(cnt) as cnt,max(last_dt) as last_dt from
(
   select concat_ws('_', cast(ceiling(rand()*99) as string), cookie_id) as tkey,event_query, count(*) as cnt,
 max(from_unixtime(unix_timestamp(event_time), 'yyyyMMddHHmmss')) as last_dt
        from raw_query
 where dt >= '20220718' and dt <= '20221019' and event_query is not null
     group by concat_ws('_', cast(ceiling(rand()*99) as string), cookie_id),event_query
) temp
group by split(tkey,'_')[1],event_query;