hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "M. C. Srivas" <mcsri...@gmail.com>
Subject Re: HBase Stability
Date Mon, 21 Mar 2011 21:24:22 GMT
You will need about 2G per datanode jvm and 4G per region-server jvm
process. Plus another 500-750M for the operating system itself.

That 2-4G DRAM per machine is too low for HBase and the Datanode to function
well together (2G is extremely low to the point of unusable). I am assuming
of course that you are running the regionserver on the datanode.


On Mon, Mar 21, 2011 at 2:17 PM, Stuart Scott <Stuart.Scott@e-mis.com>wrote:

> Hi David,
>
> Thanks for your email. We have tried this fix from MSLAB's. It didn't
> make any difference. I think it is probably down to our lack of
> knowledge and our configuration. We did try some of the CDH builds from
> Cloudera and had varying results.
>
> What other information could I provide? The servers are 4-5 years old,
> just running on IDE 750gb drives (for testing) with 2-4 gb of RAM in
> each of the 16 servers.
> We only have a 100MB network at present.
>
> Should Hbase work on this level of hardware? Up to 100 million rows. We
> are obviously happy for it to take a while but it shouldn't collapse?
>
> Regards
>
> Stuart
>
> -----Original Message-----
> From: Buttler, David [mailto:buttler1@llnl.gov]
> Sent: 21 March 2011 20:46
> To: user@hbase.apache.org
> Subject: RE: HBase Stability
>
> Have you seen Todd Lipcon's post on MSLAB's?
> http://www.cloudera.com/blog/2011/02/avoiding-full-gcs-in-hbase-with-mem
> store-local-allocation-buffers-part-1/
>
> This is a new feature in 0.90.1 that prevents memory fragmentation
> during write loads.  You do have to explicitly enable this, but it seems
> a likely culprit for unstable writes
>
> But, of course, this is just a shot in the dark given the limited
> information you have provided.
>
> Dave
>
> -----Original Message-----
> From: Ted Dunning [mailto:tdunning@maprtech.com]
> Sent: Monday, March 21, 2011 1:19 PM
> To: user@hbase.apache.org
> Cc: Stuart Scott
> Subject: Re: HBase Stability
>
> Is there a reason you are not using a recent version of 0.90?
>
> On Mon, Mar 21, 2011 at 1:17 PM, Stuart Scott
> <Stuart.Scott@e-mis.com>wrote:
>
> > We are using Hbase 0.89.20100924+28, r
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message