I'm basically referring to federating multiple namenodes (connecting two different hdfs instances under a single namespace so data can be distributed across them). Here's the documentation for Hadoop 2.6.0 [1]
[1] https://hadoop.apache.org/docs/r2.6.0/hadoop-project-dist/hadoop-hdfs/Federation.html On Fri, Sep 25, 2015 at 12:42 AM, Ashish Kumar9 <ashis...@in.ibm.com> wrote: > This is interesting . Can you share any blog/document that talks > multi-volume HDFS instances . > > Thanks and Regards, > Ashish Kumar > > > From: Corey Nolet <cjno...@gmail.com> > To: user@hadoop.apache.org > Date: 09/24/2015 10:40 PM > Subject: Re: Hetergeneous Hadoop Cluster > ------------------------------ > > > > If the hardware is drastically different, I would think a multi-volume > HDFS instance would be a good idea (put like-hardware in the same volumes). > > On Mon, Sep 21, 2015 at 3:29 PM, Tushar Kapila <*tgkp...@gmail.com* > <tgkp...@gmail.com>> wrote: > Would only matter if OS specific communication was being used between > nodes. I assume they do not do that. > If that is true -> It would depend on the network and each nodes config > for the work it is doing. Cluster performance would not suffer just because > it is heterogeneous. > > On Mon, Sep 21, 2015 at 8:09 PM, Ashish Kumar9 <*ashis...@in.ibm.com* > <ashis...@in.ibm.com>> wrote: > Hi : > > Has anyone tried a heterogeneous hadoop cluster with management nodes and > data nodes running on multiple linux distros. and on multiple h/w > architecture . > > If so , what is the performance of such cluster . > > Thanks > Ashish > > > > -- > Regards > Tushar Kapila > >
