The speed shown there is to be taken with a grain of salt. It is an
average value measured from the initiation of its phase. So if the
reduce is waiting for more map outputs to be available, having started
earlier (default is at 5% of maps completed) then the wait period is
also counted into this rate and thereby drops down in value even
though the real copy speed may be higher. A better way to check would
be to simply measure the network traffic if you want to.
On Thu, Feb 28, 2013 at 12:53 PM, Patai Sangbutsarakum
<[EMAIL PROTECTED]> wrote:
> Thanks Harsh, you always the first..
> Yeah, that's really make sense, copy inbound those output of mappers
> to running reduce task attempt.
> I am trying to think that the speed of 0.44MB/s is pretty low to me.
> i am deciding if it is because of data is not that much to copy as
> because not all the mappers are finished at the same time.
> or it is the problem of the network itself. (i already check that bond0 is 1gb)
> On Wed, Feb 27, 2013 at 11:06 PM, Harsh J <[EMAIL PROTECTED]> wrote:
>> The latter (from other machines, inbound to where the reduce is
>> running, onto the reduce's local disk, via mapred.local.dir). The
>> reduce will, obviously, copy outputs from all maps that may have
>> produced data for its assigned partition ID.
>> On Thu, Feb 28, 2013 at 12:27 PM, Patai Sangbutsarakum
>> <[EMAIL PROTECTED]> wrote:
>>> Good evening Hadoopers!
>>> at the jobtracker page, click on a job, and click at running reduce
>>> task, I am going to see
>>> task_201302271736_0638_r_000000 reduce > copy (136 of 261 at 0.44 MB/s)
>>> I am really curious where is the data is being copy.
>>> if i clicked at the task, it will show a host that is running the task attempt.
>>> question is "reduce > copy" is referring data copy outbound from host
>>> that is running task attempt, or
>>> referring to data is being copy from other machines inbound to this
>>> host (that's running task attempt)
>>> and in both cases how do i know what machines that host is copy data from/to?
>> Harsh J