Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hadoop >> mail # user >> Re: 答复: Why hadoop is spawing two map over file size 1.5 KB ?


Copy link to this message
-
Re: 答复: Why hadoop is spawing two map over file size 1.5 KB ?
Hi

We just have discussed about the similar problem(the title is Increase the number of mappers in PM mode),as I thought.

First of all you can not setup the the number of mappers.however you change the mapred.max.split.size, or mapred.min.split.size ,the number of the mapper will change automatically.

I think the two mappers which you got in your situation,that is because the value of the mapred.max.split.size, or mapred.min.split.size.
 
Do please check that .and appreciate that you give the results.
Regards.

发自我的 iPhone

在 2013-3-15,0:04,王志强 <[EMAIL PROTECTED]> 写道:

> please check parameter "mapred.map.tasks" is equals to 2?
> 发件人: Yanbo Liang [[EMAIL PROTECTED]]
> 发送时间: 2013年3月14日 23:11
> 收件人: [EMAIL PROTECTED]
> 主题: Re: Why hadoop is spawing two map over file size 1.5 KB ?
>
> I guess may be one of them is the speculative execution.
> You can check the parameter "mapred.map.tasks.speculative.execution" to ensure whether it is allowed speculative execution.
> You can get the precise information that whether it is speculative map task from the tasktracker log.
>
> 2013/3/12 samir das mohapatra <[EMAIL PROTECTED]>
>> Hi All,
>>   I have very fundamental doubt, I have file having size 1.5KB and block size is default block size, But i could see two mapper it got creted during the Job. Could you please help  to get whole picture why it is .
>>
>> Regards,
>> samir.
>
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB