hadoop-common-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Phantom <ghostwhoowa...@gmail.com>
Subject Re: Writes to HDFS
Date Tue, 07 Aug 2007 22:16:02 GMT
Since data doesn't exist till the file is closed, I for write for an hour
and then close the file. Now does every write goes to the namenode and then
to the datanode ? Reason I ask this is because in an hour I write on an
average 5GB of data which means multiple blocks and hence the data may be
placed on different datanodes. Basically my question is whether the namenode
is involved in every write request.

Thanks
A

On 8/7/07, Raghu Angadi <rangadi@yahoo-inc.com> wrote:
>
>
> To write data to HDFS, you should be able to connect to both Namenode
> and to all of the datanodes. NameNode decides where to place the data
> blocks and tells the client (you) to send the blocks to corresponding
> datanodes.
>
> Raghu.
>
> Phantom wrote:
> > Hi
> >
> > When I write data into HDFS do I always need to connect to the datanode
> to
> > write the data ? Can I connect to any namenode to do so ? If so how does
> the
> > datanode keep track of where the various chunks are going to be placed ?
> Any
> > resource that I can look at or could someone explain this to me ?
> >
> > Thanks
> > A
> >
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message