hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jean-Daniel Cryans <jdcry...@apache.org>
Subject Re: Zookeeper quorum config
Date Wed, 08 Dec 2010 18:24:19 GMT
Ted,

Sorry, wrong choice of words, HBase will be unreliable. I'm referring
to a situation where the session timeout is caused by a very slow
quorum because, as I saw it happening before, the datanodes where
pegging the disk(s) while being hammered by the region servers.

J-D

On Wed, Dec 8, 2010 at 10:20 AM, Ted Dunning <tdunning@maprtech.com> wrote:
> This is a bit misleading.  ZK is always reliable regardless of disk latency.
>  All that happens on a busy disk is that
> you get longer latency for ZK transactions.  For a dedicated and
> well-configured machine, you can have average
> latency (including committing to disk) of about 7 ms.  For a multi-purpose
> busy machine, you may see latencies
> of 300 ms.
>
> Neither case will cause unreliable operation.
>
> On Wed, Dec 8, 2010 at 10:14 AM, Jean-Daniel Cryans <jdcryans@apache.org>wrote:
>
>> Basically, ZK simply needs the lowest latency to disk and network in
>> order to work reliably. It's not CPU intensive, and it's only memory
>> intensive if you are using tons of znodes (HBase doesn't).
>>
>

Mime
View raw message