I think "having one such "slow" RS will make the whole cluster work slower
(basically, at its speed)." is not 100% accurate.

The "slowness" usually affects HDFS and ripples into HBase in many
different ways. I've seen in some cases  the DN process is not started and
HBase needs to get all the blocks from the network and that affects all
operations on that RS. In other cases is just an unresponsive drive that
causes the HDFS pipeline to be slow for some clients and that causes write
operations to randomly stall (due appending to the WAL) for few seconds
across the cluster, and in some other cases the NIC speed flips to 100mbits
causing all kind of issues (compactions or the memstore flushing taking
longer).

esteban.


--
Cloudera, Inc.



On Thu, Apr 3, 2014 at 12:28 PM, yonghu <[email protected]> wrote:

> I think the right understanding of this is it will slow down the data query
> processing. You can think the RS who hit a heady I/O as a hotspot node. It
> will not slow down the whole cluster, it will only slow down the data
> applications which access the data from that RS.
>
>
> On Thu, Apr 3, 2014 at 3:58 AM, Libo Yu <[email protected]> wrote:
>
> > "having one such "slow" RS will make the whole cluster work slower
> > (basically, at its speed)."
> > http://blog.sematext.com/2012/07/16/hbase-memstore-what-you-should-know/
> >
> > From: [email protected]
> > To: [email protected]
> > Subject: single node's performance and cluster's performance
> > Date: Wed, 2 Apr 2014 09:06:35 -0400
> >
> >
> >
> >
> > Hi all,
> >
> > I read an  article yesterday about Hbase. It says if a region server
> > has a performance hit by heavy IO, that will impact the whole cluster.
> > So I want to know how a single node's performance issue will slow
> > down all the cluster. Thanks.
> >
> > Libo
> >
> >
> >
>

Reply via email to