| | 1 | {{{ |
| | 2 | #!html |
| | 3 | <div style="text-align: center;"><big |
| | 4 | style="font-weight: bold;"><big><big>實做四: Hadoop 程式編譯</big></big></big></div> |
| | 5 | }}} |
| | 6 | [[PageOutline]] |
| | 7 | |
| | 8 | |
| | 9 | == 前言:啟動Hadoop環境 == |
| | 10 | |
| | 11 | * 重新啟動昨天的環境 |
| | 12 | * 在 node1 上操作 |
| | 13 | {{{ |
| | 14 | $ cd ~ |
| | 15 | $ wget http://hadoop.nchc.org.tw/~waue/clean.sh |
| | 16 | $ chmod 755 clean.sh |
| | 17 | $ ./clean.sh |
| | 18 | $ |
| | 19 | }}} |
| | 20 | * 請檢查 hadoop 是否正確運作. |
| | 21 | |
| | 22 | == 練習 1 : Word Count 初級版 == |
| | 23 | |
| | 24 | * 上傳內容到hdfs內 |
| | 25 | |
| | 26 | {{{ |
| | 27 | $ cd /opt/hadoop |
| | 28 | $ bin/hadoop fs -mkdir input |
| | 29 | $ echo "I like NCHC Cloud Course." > input1 |
| | 30 | $ echo "I like nchc Cloud Course, and we enjoy this course." > input2 |
| | 31 | $ bin/hadoop fs -put input1 input |
| | 32 | $ bin/hadoop fs -put input2 input |
| | 33 | $ bin/hadoop fs -ls input |
| | 34 | }}} |
| | 35 | |
| | 36 | * 點此連結 [attachment:wiki:jazz/Hadoop_Lab6:WordCount.java?format=raw WordCount.java] 並將他存到 /opt/hadoop; |
| | 37 | |
| | 38 | |
| | 39 | * 運作程式 |
| | 40 | |
| | 41 | {{{ |
| | 42 | $ mkdir MyJava |
| | 43 | $ javac -classpath hadoop-*-core.jar -d MyJava WordCount.java |
| | 44 | $ jar -cvf wordcount.jar -C MyJava . |
| | 45 | $ bin/hadoop jar wordcount.jar WordCount input/ output/ |
| | 46 | $ bin/hadoop fs -cat output/part-00000 |
| | 47 | }}} |
| | 48 | |
| | 49 | ----- |
| | 50 | |
| | 51 | == 練習 2 : Word Count 進階版 == |
| | 52 | |
| | 53 | {{{ |
| | 54 | $ echo "\." >pattern.txt && echo "\," >>pattern.txt |
| | 55 | $ bin/hadoop fs -put pattern.txt ./ |
| | 56 | $ mkdir MyJava2 |
| | 57 | }}} |
| | 58 | |
| | 59 | * 點此連結 [attachment:wiki:jazz/Hadoop_Lab6:WordCount2.java?format=raw WordCount2.java] 並將他存到 /opt/hadoop; |
| | 60 | |
| | 61 | {{{ |
| | 62 | $ javac -classpath hadoop-*-core.jar -d MyJava2 WordCount2.java |
| | 63 | $ jar -cvf wordcount2.jar -C MyJava2 . |
| | 64 | $ bin/hadoop jar wordcount2.jar WordCount2 input output2 -skip pattern.txt |
| | 65 | $ bin/hadoop fs -cat output2/part-00000 |
| | 66 | $ bin/hadoop jar wordcount2.jar WordCount2 -Dwordcount.case.sensitive=false input output3 -skip pattern.txt |
| | 67 | $ bin/hadoop fs -cat output3/part-00000 |
| | 68 | }}} |