下面是nutch fetch job中map shuffle reduce的时间花费的一个列表:
server name | Fri Mar 05 09:45:13 GMT 2010 | job_201003050945_0006 | fetch crawl/segments/20100305102846 | user name |
User : username --用户名
JobName : fetch crawl/segments/20100305102846 --job的名称
JobConf : hdfs://servername:9000/opt/crawler/data/mapred/system /job_201003050945_0006/job.xml --使用的配置文件位置
Submitted At : 5/03 10:30:29 --提交时间
Launched At : 5/03 10:30:30 (0sec) --开始时间
Finished At : 6/03 17:04:09 (30hrs, 33mins, 38sec) --结束时间
Status : SUCCESS --结束状态
---从下面的分析可以得出map时间 22hrs avg
---从下面的分析可以得出shuffle时间 30hrs avg
---从下面的分析可以得出reduce时间 29mins avg
Average time taken by Map tasks: 22hrs, 6mins, 40sec
Worse performing map tasks
Task Id | Time taken |
task_201003050945_0006_m_000010 | 24hrs, 47mins, 14sec |
task_201003050945_0006_m_000011 | 24hrs, 44mins, 1sec |
task_201003050945_0006_m_000013 | 24hrs, 42mins, 23sec |
task_201003050945_0006_m_000012 | 24hrs, 29mins, 6sec |
task_201003050945_0006_m_000007 | 24hrs, 19mins, 44sec |
task_201003050945_0006_m_000006 | 24hrs, 18mins, 54sec |
task_201003050945_0006_m_000001 | 24hrs, 18mins, 41sec |
task_201003050945_0006_m_000008 | 24hrs, 18mins, 26sec |
task_201003050945_0006_m_000000 | 24hrs, 17mins, 7sec |
task_201003050945_0006_m_000005 | 24hrs, 16mins, 2sec |
The last Map task task_201003050945_0006_m_000016 finished at (relative to the Job launch time): 6/03 16:32:44 (30hrs, 2mins, 14sec)
Average time taken by Shuffle: 30hrs, 2mins, 10sec
Worse performing Shuffle(s)
Task Id | Time taken |
task_201003050945_0006_r_000000 | 30hrs, 2mins, 26sec |
task_201003050945_0006_r_000002 | 30hrs, 2mins, 18sec |
task_201003050945_0006_r_000001 | 30hrs, 2mins, 18sec |
task_201003050945_0006_r_000003 | 30hrs, 2mins, 4sec |
task_201003050945_0006_r_000005 | 30hrs, 2mins, 3sec |
task_201003050945_0006_r_000006 | 30hrs, 2mins, 2sec |
task_201003050945_0006_r_000004 | 30hrs, 2mins, 0sec |
The last Shuffle task_201003050945_0006_r_000000 finished at (relative to the Job launch time): 6/03 16:33:08 (30hrs, 2mins, 37sec)
Average time taken by Reduce tasks: 29mins, 38sec
Worse performing reduce tasks