不多说,直接上干货!
前期博客
hadoop-2.6.0.tar.gz + spark-1.6.1-bin-hadoop2.6.tgz + zeppelin-0.5.6-incubating-bin-all.tgz(master、slave1和slave2)(博主推荐)(图文详解)
hadoop-2.7.3.tar.gz + spark-2.0.2-bin-hadoop2.7.tgz + zeppelin-0.6.2-incubating-bin-all.tgz(master、slave1和slave2)(博主推荐)(图文详解)
测试
配置Spark解释器
创建Note
运行官网的例子:
然后我们点Notebook->Zeppelin Tutorial
然后点击save
然后在最上面的部分点击如下图所示的位置:
运行结果在截取两张图给大家:
最后一个要注意的是,这是我自己摸索着搞得,有问题发生是正常的,所以标题加上了初探两个字!
ml-100k推荐数据集的下载
官网
https://grouplens.org/datasets/movielens/
因为,这里,考虑到测试入门而已。只要跑通,数据量大小其次。所以,选择下载ml-100k。
[hadoop@master data]$ pwd
/home/hadoop/data
[hadoop@master data]$ ll
total
drwxrwxr-x hadoop hadoop Apr : es
drwxrwxr-x hadoop hadoop Apr : flume
drwxrwxr-x hadoop hadoop May : jar
drwxrwxr-x hadoop hadoop Jul : storm
drwxrwxr-x. hadoop hadoop Apr : tmp
drwxrwxr-x. hadoop hadoop Apr : zookeeper
[hadoop@master data]$ mkdir zeppelin
[hadoop@master data]$ ll
total
drwxrwxr-x hadoop hadoop Apr : es
drwxrwxr-x hadoop hadoop Apr : flume
drwxrwxr-x hadoop hadoop May : jar
drwxrwxr-x hadoop hadoop Jul : storm
drwxrwxr-x. hadoop hadoop Apr : tmp
drwxrwxr-x hadoop hadoop Sep : zeppelin
drwxrwxr-x. hadoop hadoop Apr : zookeeper
[hadoop@master data]$
[hadoop@master zeppelin]$ pwd
/home/hadoop/data/zeppelin
[hadoop@master zeppelin]$ ll
total
[hadoop@master zeppelin]$ wget http://files.grouplens.org/datasets/movielens/ml-100k.zip
---- ::-- http://files.grouplens.org/datasets/movielens/ml-100k.zip
Resolving files.grouplens.org... 128.101.34.235
Connecting to files.grouplens.org|128.101.34.235|:... connected.
HTTP request sent, awaiting response... OK
Length: (4.7M) [application/zip]
Saving to: “ml-100k.zip” %[=====================================================================================================================================================>] ,, 238K/s in 19s -- :: ( KB/s) - “ml-100k.zip” saved [/] [hadoop@master zeppelin]$ ll
total
-rw-rw-r-- hadoop hadoop Jan ml-100k.zip
[hadoop@master zeppelin]$
创建完成后就可以看到ml-100k页面,单击Zeppelin可以回到首页。
回到Zeppelin首页,就可以看到如下图的页面。
或者
http://192.168.80.145:8099/#/