weixin_46011754的博客

步骤一：准备日志数据数据是一个标准的JSONROW数据 加载数据步骤二：将日志数据文件导入到Hive的ODS层①：建表并导入数据create table tb_log(log string ) partitioned by(dt string);load data local inpath "/opt/data/log.log" into table tb_log partition(dt='20201007');②：json解析数据在hive中建ODS表cre

– with语法 with '1' as v select * from tb_user where id=v; --求平均年龄 with (select count(1) from tb_user) as cnt select (sum(age)/cnt) from tb_user;array join 语法array join 相当于explode+lateral viewselect id,arr from tb_arr_join array join arr; --会覆盖原先的数组

SummingMergeTree引擎建表– 将区内相同主键的所有的数据累加create table tb_summ_merge_tree(id Int8 ,name String,cDate Date,cost Float64) engine=SummingMergeTree(cost)order by id --主键partition by name;insert into tb_summ_merge_tree values(1,'zss','2020-10-05',6.9),

ClickHouse的安装1)安装curl工具 yum install -y curl2)添加clickhouse的yum镜像 curl -shttps://packagecloud.io/install/repositories/altinity/clickhouse/script.rpm.sh| sudo bash3)检查镜像情况 [root@doit04 yum.repos.d]# yum list | grepclickhouse cli