Answers inline

> -> Once I place my data in HDFS, it gets replicated and chunked
> automatically over the datanodes. Right? Hadoop takes care of all those
> things.

Yes it does

> -> Now, if there is some third party who is not participating in the Hadoop
> program. Means, he is not one of the nodes of hadoop cluster. Now, he has
> some data on his local filesystem. Thus, can I place this data into HDFS?
> How?

Using the HDFS interface you would put the data in (not dissimilar to
FTP'ing it to a server)

> -> Then, now when, that third party asks for  a file or a direcory or any
> kind of data that was previously being dumped in HDFS without that third
> person's knowledge- he wnats it back(wants to retrieve it). Thus, the data
> should get placed on his local file system again, in some specific
> directory. How can I do this?

Copy it out of HDFS onto the local file system - it is pretty much
like copying from a mounted drive to a different mounted drive, just
that you go through a hadoop command (distcp) and not a native command
(cp)

> -> Will I have to use Map-Reduce or something else ot make it work.

No.  You could write a java program or use command line utilities.
Maybe you can do it in other languages but I only do Java...

> -> Also, if I write map reduce code for all the complete activity, how will
> I fetch the data or the files that are chunked in HDFS in the form of blocks
> and combine(reassemble) them into a complete file and place it on a node;s
> local filesystem who is not a part of hadoop cluster setup.

You don't write MR code for putting and retrieving files - this is all
done by Hadoop for you.  Just copy files in and copy files out.

It's probably worth you reading the Hadoop command line guide:
http://hadoop.apache.org/core/docs/r0.19.1/commands_manual.html to get
an understanding of what you can do from the command line.  All those
command line utilities you can use programmatically (e.g. from code)
as well.

Cheers,

Tim




>
> Eagerly waiting for reply!
>
> Thanking You,
> Sugandha!
>
>
>
> --
> Regards!
> Sugandha
>

Reply via email to