【6.824】分布式lab1 mapReduce

Paper main idea

复制代码
Figure 1 shows the overall flow of a MapReduce operation in our implementation. When the user program calls the MapReduce function, the following sequence of actions occurs (the numbered labels in Figure 1 correspond to the numbers in the list below): 

1. The MapReduce library in the user program first splits the input files into M pieces of typically 16 megabytes to 64 megabytes (MB) per piece (controllable by the user via an optional parameter). It then starts up many copies of the program on a cluster of machines. 

2. One of the copies of the program is special -- the master. The rest are workers that are assigned work by the master. There are M map tasks and R reduce tasks to assign. The master picks idle workers and assigns each one a map task or a reduce task. 

3. A worker who is assigned a map task reads the contents of the corresponding input split. It parses key/value pairs out of the input data and passes each pair to the user-defined Map function. The intermediate key/value pairs produced by the Map function are buffered in memory

4. Periodically, the buffered pairs are written to local disk, partitioned into R regions by the partitioning function. The locations of these buffered pairs on the local disk are passed back to the master, who is responsible for forwarding these locations to the reduce workers. 

5. When a reduce worker is notified by the master about these locations, it uses remote procedure calls to read the buffered data from the local disks of the map workers. When a reduce worker has read all intermediate data, it sorts it by the intermediate keys so that all occurrences of the same key are grouped together. The sorting is needed because typically many different keys map to the same reduce task. If the amount of intermediate data is too large to fit in memory, an external sort is used. 

6. The reduce worker iterates over the sorted intermediate data and for each unique intermediate key encountered, it passes the key and the corresponding set of intermediate values to the user's Reduce function. The output of the Reduce function is appended to a final output file for this reduce partition

7. When all map tasks and reduce tasks have been completed, the master wakes up the user program. At this point, the MapReduce call in the user program returns back to the user code.

some ideas

Task Synchronous:

all map task should be earlier than reduce task!

some map task may wait seconds, so should first all map tasks done before reduce.

Lock:

workers run parallel, should add mutex to filelist when

Error Return:

workers may run some error, such as rpc sock connect and can not open files. When these errors happend, should return Error to master, other than just stop itself.

Heartbeat:

master should ask workers whether they are alive.

Implementation method:

  1. Set a new scheduled task and ask every 2 minutes.
  2. Every time a new worker requests a new task, add a judgment when rotating all tasks to assign tasks to the worker. If a task status is not Finish and times out, the task fails
  3. Write in the Done() method and continuously rotate to ensure that all tasks are completed. And take the opportunity to determine if there is a task timeout.

Performance comparison: 1>2>3

Judging by frequency. The execution frequency of 3 is too high, constantly executing and discussing the sequence.

status of task and worker

worker:

  • Finish: worker finish task X
  • Ready: worker ask for a new task
  • Wrong: worker meet some problems
  • Close: worker will be closed when master ask worker to close

task:

  • Run: some worker is running on this task
  • Done: this task is finished
  • (empty string): task not be allocated yet

lab test:

pass all the test!

相关推荐
玖米GEO运营王总监2 小时前
成都AI数字人视频怎么批量生产?流程与合规标注
大数据·人工智能·音视频
龙腾AI白云3 小时前
数字孪生与计算机视觉:让机器看懂现场
大数据·数据库·机器学习·知识图谱·pygame
starzy19906 小时前
Structured Streaming 案例剖析:一个 8 万 QPS 实时大屏的完整落地与三次线上事故复盘
大数据·spark
大大大大晴天16 小时前
浅聊维度建模中的星型模型与雪花模型
大数据
数字新视界20 小时前
能耗管理系统提升企业能源效率的实践与展望
大数据·人工智能·数据中心·机房管理·模块化机房
大大大大晴天21 小时前
别让金额越算越多:事实表与维度表设计避坑指南
大数据
如夏倾城1 天前
入侵检测系统(IDS)原理、部署模式与 IPS/防火墙对比
大数据·网络·人工智能·网络安全·云原生·架构·kubernetes
wuyk5551 天前
Spider-第20讲-综合实战可上线爬虫
大数据·爬虫·python·elasticsearch
培培说证1 天前
大数据与会计专业性价比高的证书有哪些?
大数据
伞伞悦读1 天前
【第90期】KNN分类回归流水线与K值对照的实验与排查
大数据·人工智能·python·深度学习·机器学习