Really, for 5 nodes, the master is your SPOF, so unless you plan to
use ZK for anything else then it's probably going to be easier/faster
to have only 1 ZK server. Collocating ZK and region servers is risky
if ZK doesn't have it's own disk, doing it 3 times is 3 times worse :)
Also it may save you
I have master also serving as regionserver. I'll run ZK also on 3 of the
regionservers. I don't have too much data (few TBs only), so I guess it
would be fine?
On Thu, Dec 9, 2010 at 12:44 AM, Ted Dunning wrote:
> Ahh... that is very much at the other end of the spectrum from what I am
> used t
Ahh... that is very much at the other end of the spectrum from what I am
used to.
Yes. It would not be good to run ZK on a system where the disk is
essentially unavailable for
significant amounts of time.
On Wed, Dec 8, 2010 at 10:24 AM, Jean-Daniel Cryans wrote:
> Ted,
>
> Sorry, wrong choice
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 a
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-pur
ZK should have a separate disk IF you have tight latency or high throughput
requirements.
If ZK committing the log after other disk writes is OK and if you are doing
dozens of transactions
per second rather than , then you are fine sharing a disk. In terms of
latency, with a shared and busy
disk
On 5 machines, put the standalone ZK with the master since the disk's
usage is slow (same for namenode).
For some HW info, see this page and scroll down to the zookeeper
requirements
http://people.apache.org/~stack/hbase-0.90.0-candidate-1/docs/notsoquick.html#requirements
Basically, ZK simply n
Hi
why is it generally said that zk should have a separate disk of it's
own? Currently in my small 5 node cluster, one of the regionservers
itself is also the zk. How beneficial is it to move it to a separate
machine? And in that case, what should be the machine config
like?should it be ram intens
