环境信息：CarbonData Carbon Thrift Server集成OSS - 环境搭建

1、运行carbondata thrift server

/home/carbondata/spark-2.2.1-bin-hadoop2.7/bin/spark-submit --class org.apache.carbondata.spark.thriftserver.CarbonThriftServer $SPARK_HOME/carbonlib/apache-carbondata-1.6.0-SNAPSHOT-bin-spark2.2.1-hadoop2.7.2.jar <table-path-on-s3> <access-key> <secret-key> <s3-endpoint>

其中access-key，secret-key，s3-endpoint，table-path-on-s3为CarbonData oss bucket管理中提到的4个基本要素

6、使用Beeline连接thrift server

1. Beeline命令

cd $SPARK_HOME
./bin/beeline -u jdbc:hive2://localhost:10000

2. Beeline中输入创建表语句
   CREATE TABLE IF NOT EXISTS test_table ( id string, name string,city string,age Int) STORED AS carbondata LOCATION 's3a://demo20190203/carbon/session/data/store' ;

oss中数据如下：

image.png

3）Beeline中载入数据到OSS
LOAD DATA INPATH 'hdfs://localhost:9000/tmp/sample.csv' INTO TABLE test_table;

OSS中数据如下：

image.png
image.png

4)Beeline中查询数据
SELECT city, avg(age), sum(age) FROM test_table GROUP BY city;
查询的数据如下：

image.png