It's unsynchronized however it's creating a new BlockReader on each
call, that seems like a problem?
at 5:58 PM, Todd Lipcon wrote:
> Hi Jason,
>
> Yes, this method is currently very inefficient.. HDFS-941 will
> hopefully improve this situation, but currently there's no
> particularly efficient way to do multithreaded random access.
>
> -Todd
>
> On Tue, Mar
gt; something that I won't be supporting myself in production against an
> increasingly diverging upstream.
>
> - Andy
>
>
> --- On Wed, 3/2/11, Todd Lipcon wrote:
>
>> From: Todd Lipcon
>> Subject: Re: Is DFSInputStream.read(long position,...) designed
This is an interesting question. AvatarNode doesn't use Zookeeper?
Is there a drawback to implementing ZK for the NameNode other than the
potential bugs from a newly engineered feature?
Perhaps this issue is the main one for implementing the NameNode in
ZK? HDFS-1623
On Fri, Apr 22, 2011 at 10:
ally it was not using ZooKeeper. That is way I raised that
> question. Also, AvatorNode is build on HDFS 0.20 but my intension is to
> utilize BackupNode that comes with HDFS 0.21
>
> Thanks,
> Ozcan.
>
> -Original Message- From: Jason Rutherglen
> Sent: Sat
Is there a way to send a request to the name node to replicate
block(s) to a specific DataNode? If not, what would be a way to do
this? -Thanks
e nice to expose this functionality as a "limited
> private evolving" API.
>
> In general, though, keep in mind that, whenever you write data, you'll
> get a local copy first, if the writer is in the cluster. That's how
> HBase gets locality for most of its accesse
y 26, 2011 at 12:30 PM, Todd Lipcon wrote:
> On Thu, May 26, 2011 at 12:02 PM, Jason Rutherglen
> wrote:
>> Todd, thanks!
>>
>>> In general, though, keep in mind that, whenever you write data, you'll
>>> get a local copy first, if the writer is in the clus