hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From anilkr <kr_an...@rediffmail.com>
Subject Re: Basic question about using C# with Hadoop filesystems
Date Sun, 10 Jan 2010 15:17:08 GMT

Aram, where is the discussion about fuse-dfs, could not find the link in your


Aram Mkhitaryan wrote:
> here is a discussion with subject 'fuse-dfs',
> there they discuss problems with mounting hdfs,
> probably you can ask your question there
> On Sun, Jan 10, 2010 at 7:06 PM, Aram Mkhitaryan
> <aram.mkhitaryan@googlemail.com> wrote:
>> I'm not an expert here, but
>> I read somewhere that it's possible to install a module in linux
>> which will allow you to mount hdfs folder as a standard linux folder
>> if I'm not wrong it was in one of the claudera's distributions,
>> probably you can find something there
>> On Sun, Jan 10, 2010 at 5:42 PM, anilkr <kr_anil2@rediffmail.com> wrote:
>>> Thank you Ryan,
>>> My C# code is also on Linux (as it uses MONO framework on Linux).
>>> I understand that some bridging would be required. I am thinking that
>>> since
>>> Hadoop exposes some C APIs to operate with the Hadoop filesystem. If i
>>> write
>>> a wrapper in C and make a DLL out of it. Now my C# application will call
>>> this DLL for file operations and this DLL will call Hadoop APIs to
>>> operate
>>> with the files.
>>> Do you think this would be a proper way.
>>> thanks again
>>> Ryan Rawson wrote:
>>>> Hadoop fs is not a typical filesystem, it is rpc oriented and uses a
>>>> thick
>>>> client in Java. To get access to it from c# would involve bridging to
>>>> Java
>>>> somehow. The c++ client does this.
>>>> Most of hbase devs use Mac or Linux boxes. We aren't really experts in
>>>> windows tech. Maybe the main hadoop list could help you?
>>>> On Jan 9, 2010 11:50 PM, "anilkr" <kr_anil2@rediffmail.com> wrote:
>>>> Currently my application uses C# with MONO on Linux to communicate to
>>>> local
>>>> file systems (e.g. ext2, ext3). The basic operations are open a file,
>>>> write/read from file and close/delete the file. For this currently i
>>>> use
>>>> C#
>>>> native APIs to operate on the file.
>>>>   My Question is: If i install Hadoop file system on my Linux box. Then
>>>> what change i need to do to my existing functions so that they
>>>> communicate
>>>> to hadoop file system to do basic operations on the file. Since Hadoop
>>>> infrastructure is based on Java, How any C# (with MONO) application
>>>> will
>>>> do
>>>> basic operations with Hadoop. Do the basic APIs in C# to operate on a
>>>> file
>>>> (likr File.Open or File.Copy) work well with Hadoop filesystems too?
>>>> Also, If i want to open a file then do i need to mount to Hadoop
>>>> filesystem
>>>> programmatically? If yes, how?
>>>> Thanks,
>>>> Anil
>>>> --
>>>> View this message in context:
>>>> http://old.nabble.com/Basic-question-about-using-C--with-Hadoop-filesystems-tp27096203p27096203.html
>>>> Sent from the HBase User mailing list archive at Nabble.com.
>>> --
>>> View this message in context:
>>> http://old.nabble.com/Basic-question-about-using-C--with-Hadoop-filesystems-tp27096203p27098395.html
>>> Sent from the HBase User mailing list archive at Nabble.com.

View this message in context: http://old.nabble.com/Basic-question-about-using-C--with-Hadoop-filesystems-tp27096203p27099222.html
Sent from the HBase User mailing list archive at Nabble.com.

View raw message