hadop-2.6.0总结
1. 下载（略）
2. 编译（略）
3. 配置（伪分布、集群略）
4. Hdfs
1. Web interface:http://namenode-name:50070/(显示datanode列表和集群统计信息)
2. shell command & dfsadmin comman
3. checkpoint node & backup node
1. fsimage和edits文件merge原理
2. （猜测是早期版本的特性）手动恢复宕掉的集群：import checkpoint；
3. backupnode: Backup Node在内存中维护了一份从Namenode同步过来的fsimage，同时它还从namenode接收edits文件的日志流，并把它们持久化硬盘，Backup Node把收到的这些edits文件和内存中的fsimage文件进行合并，创建一份元数据备份。Backup Node高效的秘密就在这儿，它不需要从Namenode下载fsimage和edit，把内存中的元数据持久化到磁盘然后进行合并即可。
4. banlancer:平衡各rock和datanodes数据不均衡
5. Rock awareness：机架感知
6. Safemode：当数据文件不完整或者手动进入safemode时，hdfs只读，当集群检查达到阈值或手动离开安全模式时，集群恢复读写。
7. Fsck：块文件检查命令
8. Fetchdt:获取token（安全）
9. Recovery mode:恢复模式
10. Upgrade and Rollback：升级、回滚
11. File Permissions and Security
12. Scalability
13.
5. Mapreduce
1.
public class MyMapper extends Mapper<Object, Text, Text, IntWritable>{
private Text word = new Text();
private IntWritable one = new IntWritable(1);
// 重写map方法
@Override
public void map(Object key, Text value, Context context)
throws IOException, InterruptedException {
StringTokenizer stringTokenizer = new StringTokenizer(value.toString());
while(stringTokenizer.hasMoreTokens()){
word.set(stringTokenizer.nextToken());
// （word,1）进行传递
context.write(word, one);
}
}
}
public class MyReducer extends Reducer<Text, IntWritable, Text, IntWritable>{
private IntWritable result = new IntWritable(0);
// 重写reduce方法
@Override
protected void reduce(Text key, Iterable<IntWritable> iterator,
Context context) throws IOException, InterruptedException {
int sum = 0;
for(IntWritable i : iterator){
sum += i.get();
}
result.set(sum);
// reduce输出的值
context.write(key, result);
}
}
public class WordCountDemo {
public static void main(String[] args) throws Exception {
Configuration conf = new Configuration();
Job job = Job.getInstance(conf, "word count");
job.setJarByClass(WordCountDemo.class);
// 设置map、reduce class
job.setMapperClass(MyMapper.class);
job.setReducerClass(MyReducer.class);
job.setCombinerClass(MyReducer.class);
// 设置最终输出的格式
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(IntWritable.class);
// 设置FileInputFormat outputFormat
FileInputFormat.addInputPath(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
System.exit(job.waitForCompletion(true) ? 0 : 1);
}
}
2. Job.setGroupingComparatorClass(Class).
3. Job.setCombinerClass(Class),
4. CompressionCodec
5. Map数：Configuration.set(MRJobConfig.NUM_MAPS, int) => dataSize/blockSize
6. Reducer数：Job.setNumReduceTasks(int).
With 0.95 all of the reduces can launch immediately and start transferring map outputs as the maps finish. With 1.75 the faster nodes will finish their first round of reduces and launch a second wave of reduces doing a much better job of load balancing.
7. Reduce->shuffle: Input to the Reducer is the sorted output of the mappers. In this phase the framework fetches the relevant partition of the output of all the mappers, via HTTP. –> reduce是mapper排序后的输出的结果。在这一阶段，框架通过http抓取所有mapper输出的有关分区。
8. Reduce ->sort：The framework groups Reducer inputs by keys (since different mappers may have output the same key) in this stage. The shuffle and sort phases occur simultaneously; while map-outputs are being fetched they are merged.-> 在这一阶段，框架按照输入的key（不同的mapper可能输出相同的key）分组reducer。Shuffle和sort会同时发生，当map输出被捕捉时，他们又会进行合并。
9. Reduce ->reduce：
10. Secondary sort
11. Partitioner
12. Counter ：Mapper and Reducer implementations can use the Counter to report statistics.
13. Job conf：配置 -> speculative manner ( setMapSpeculativeExecution(boolean))/setReduceSpeculativeExecution(boolean)), maximum number of attempts per task (setMaxMapAttempts(int)/ setMaxReduceAttempts(int)) etc.
Or
Configuration.set(String, String)/ Configuration.get(String)
14. Task executor & environment -> The user can specify additional options to the child-jvm via the mapreduce.{map|reduce}.java.opts and configuration parameter in the Job such as non-standard paths for the run-time linker to search shared libraries via -Djava.library.path=<> etc. If the mapreduce.{map|reduce}.java.opts parameters contains the symbol @taskid@ it is interpolated with value of taskid of the MapReduce task.
15. Memory management - > Users/admins can also specify the maximum virtual memory of the launched child-task, and any sub-process it launches recursively, using mapreduce.{map|reduce}.memory.mb. Note that the value set here is a per process limit. The value for mapreduce.{map|reduce}.memory.mb should be specified in mega bytes (MB). And also the value must be greater than or equal to the -Xmx passed to JavaVM, else the VM might not start.
16. Map Parameters ...... (http://hadoop.apache.org/docs/r2.6.0/hadoop-mapreduce-client/hadoop-mapreduce-client-core/MapReduceTutorial.html#MapReduce_Tutorial)
17. Parameters ()
18. Job submission and monitoring:
1.Job provides facilities to submit jobs, track their progress, access component-tasks' reports and logs, get the MapReduce cluster's status information and so on.
2. The job submission process involves:
1. Checking the input and output specifications of the job.
2. Computing the InputSplit values for the job.
3. Setting up the requisite accounting information for the DistributedCache of the job, if necessary.
4. Copying the job's jar and configuration to the MapReduce system directory on the FileSystem.
5. Submitting the job to the ResourceManager and optionally monitoring it's status.
3. Job history
19. Job controller
1. Job.submit() || Job.waitForCompletion(boolean)
2. 多Mapreduce job
1. 迭代式mapreduce（上一个mr作为下一个mr的输入，缺点：创建job对象的开销、本地磁盘读写io和网络开销大）
2. MapReduce-JobControl：job封装各个job的依赖关系，jobcontrol线程管理各个作业的状态。
3. MapReduce-ChainMapper/ChainReduce：（chainMapper.addMap().可以在一个job中链接多个mapper任务，不可用于多reduce的job）。
20. Job input & output
1. InputFormat TextInputFormat FileInputFormat
2. InputSplit FileSplit
3. RecordReader
4. OutputFormat OutputCommitter
Yarn