Hive擴展功能(一)--Parquet

軟件環境:

linux系統: CentOS6.7
Hadoop版本: 2.6.5
zookeeper版本: 3.4.8

主機配置:

一共m1, m2, m3這五部機, 每部主機的用戶名都爲centos

192.168.179.201: m1 
192.168.179.202: m2 
192.168.179.203: m3 

m1: Zookeeper, Namenode, DataNode, ResourceManager, NodeManager, Master, Worker
m2: Zookeeper, Namenode, DataNode, ResourceManager, NodeManager, Worker
m3: Zookeeper, DataNode, NodeManager, Worker

資料:

官方資料:
    https://cwiki.apache.org/confluence/display/Hive/Parquet

注意:

Hive擴展Parquet有如下兩種方案:(hive自0.13版本之後已自帶paruqet功能, 沒必要再擴展parquet功能)

---

1.方案一:

(1) 下載Parquet擴展包

經過yum下載parquet資源包 (此方案大多數狀況下沒法找到parquet, 緣由是沒有配置YUM源, 如何配置YUM源請自行查找資料)

sudo yum -y install parquet

(2)將下載好的擴展包拷貝到指定位置

將下載到/usr/lib/parquet目錄下的jar(除javadoc.jar和sources.jar外)拷貝到$HIVE_HOME/lib目錄下

cp  /usr/lib/parquet/*  $HIVE_HOME/lib

---

2.方案二:

(1)打開$HIVE_HOME/lib目錄

cd  /home/centos/soft/hive/lib

(2)下載Parquet資源包

執行如下指令, 下載parquet資源包(注意:parquet1.2.5版本對應Hive 0.12版本,每一個版本的Hive對應的Parquet版本不一樣, 具體Hive所對應的版本請自行下載源碼查看)

for f in parquet-avro parquet-cascading parquet-column parquet-common parquet-encoding parquet-generator parquet-hadoop parquet-hive parquet-pig parquet-pig-bundle parquet-scrooge parquet-test-hadoop2 parquet-thrift
do
    curl -O https://oss.sonatype.org/service/local/repositories/releases/content/com/twitter/${f}/1.2.5/${f}-1.2.5.jar
    curl -O https://oss.sonatype.org/service/local/repositories/releases/content/com/twitter/parquet-format/1.2.5/parquet-format-1.2.5.jar
done