hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Patrick Hunt <ph...@apache.org>
Subject Re: Use of SSDs wuz RE: zookeeper & HBase
Date Fri, 09 Jul 2010 16:14:57 GMT

On 07/09/2010 07:16 AM, Michael Segel wrote:
> Patrick H recommends 5 ZK nodes in a production cluster. I don't disagree, but feel that
you could get away with 3 and still have a window for maintenance... however... if you can
measure you down time in terms of $100.00s per hour or more, the 5 ZKs are still pretty cheap.
(Or rather 3ZKs and then run a ZK on the name node, and a ZK on your secondary.)
> (Then you only have to build out 3 boxes, and put a second set of drives on the name
node and secondary)

Michael, that's pretty close, really it's more like "use 5 ZK servers in 
a production online serving cluster where you care about reliability". 
The beauty of ZK is that it's up to you - if you like to run 1 server 
that's fine from ZK perspective, you just won't have a high 
reliability/availability system. 3 server cluster will support 1 failure 
w/o loss of service, which is probably fine for most users (this is by 
far the most common I've seen). But if you want to be safe use 5. 5 
allows you to take 1 server out for maintenance and still be able to 
withstand a second unexpected failure w/o interruption of service.


> I haven't tried this out yet... my guess is that you probably won't see a large enough
performance increase to justify the expense...
> -Mike
> PS. I've been watching the prices of SSDs for a different application. I now see Crucial's
SATA III SSDs are actually cheaper than their SATA II drives.
>> From: aramakrishnan@languageweaver.com
>> To: user@hbase.apache.org
>> Date: Thu, 8 Jul 2010 19:38:22 -0500
>> Subject: RE: zookeeper&  HBase
>> Good to know ZK is IO intense.
>> Since ZK does not require much disk space and is IO intense. Has anyone played with
using solid state drives for ZK.
>> We have a 20 node cluster. It would be feasible to have a 3 node ZK all configured
with solid state drives.
>> Thanks
>> Arun
>> -----Original Message-----
>> From: Jonathan Gray [mailto:jgray@facebook.com]
>> Sent: Thursday, July 08, 2010 4:25 PM
>> To: user@hbase.apache.org
>> Subject: RE: zookeeper&  HBase
>> ZK is sensitive to IO starvation which is why it is recommended to keep it on a separate
node or separate disk.  In most cases, giving ZK its own disk is sufficient and dedicated
node(s) are unnecessary.
>> On smallish clusters like 10 nodes, I would recommend starting with just 1 ZK node
co-located with your NameNode and HMaster, but with a dedicated disk just for ZK.  Since the
NN is a SPOF, having one ZK doesn't really lower your fault tolerance, except that it may
be on a non-raided disk.  I encourage RAID usage for NN and ZK.  JBOD for DN/RS.
>> JG
> _________________________________________________________________
> The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with Hotmail.
> http://www.windowslive.com/campaign/thenewbusy?tile=multicalendar&ocid=PID28326::T:WLMTAGL:ON:WL:en-US:WM_HMP:042010_5

View raw message