hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vladimir Rodionov <vrodio...@carrieriq.com>
Subject RE: Hbase random read performance
Date Mon, 08 Jul 2013 17:32:58 GMT
Boris if your working set does not fit comfortably into 0.4 * 9 * 12GB ~ 43GB HBase cluster
block cache ( and I suppose
it does not, because it is less than 1% of your data set  ) - you will be depended on your
local data nodes  (OS page cache + raw disk I/O) performance.
Your maximum raw I/O is approx 150  (one HDD) * 2 (per HDDsnodes) * 9 (DN #) = 2700 IOPS -
this is what you can get in theory. You are getting 800-1000
- which is not bad taking into account that short circuit reads are probably disabled and
even not supported in your release.

1. Upgrade to the latest CDH 4.3 if possible
2. Enable short circuit reads in DNs
3. Increase number of nodes (and decrease data set size per node)
4. Increase number of disks per node
5. Increase RS heap memory and allocate more for block caches.
6. major compact critical tables to ensure RS data locality

You need to fit your working set into OS page cache  (or into block cache) to get decent random
IO performance


Best regards,
Vladimir Rodionov
Principal Platform Engineer
Carrier IQ, www.carrieriq.com
e-mail: vrodionov@carrieriq.com

________________________________________
From: Ted Yu [yuzhihong@gmail.com]
Sent: Monday, July 08, 2013 5:49 AM
To: user@hbase.apache.org
Subject: Re: Hbase random read performance

Moving to HBase user mailing list.

Can you upgrade to newer release such as 0.94.8 ?

Cheers

On Jul 8, 2013, at 4:36 AM, Boris Emelyanov <emelyanov@post.km.ru> wrote:

> I'm trying to configure hbase for fully random read performance, my cluster parameters
are:
>
> 9 servers as slaves, each has two 1TB HDD as hadoop volumes;
> data:  800 millions 6-10 KB objects in hbase;
>
> HBase Version - 0.90.6-cdh3u5
> HBASE_HEAPSIZE=12288;
> hfile.block.cache.size = 0.4
> hfile.min.blocksize.size = 16384
> hbase.regionserver.handler.count = 100
>
> I used several recommended tuning solutions, such as
> * tunrning on bloom filters;
> * decreasing hbase block size to 16384.
>
> However, read performance is still poor, about 800-1000rps.
>
> What would you recommend to solve the problem?
>
> --
> Best regards,
>
> Boris.
>

Confidentiality Notice:  The information contained in this message, including any attachments
hereto, may be confidential and is intended to be read only by the individual or entity to
whom this message is addressed. If the reader of this message is not the intended recipient
or an agent or designee of the intended recipient, please note that any review, use, disclosure
or distribution of this message or its attachments, in any form, is strictly prohibited. 
If you have received this message in error, please immediately notify the sender and/or Notifications@carrieriq.com
and delete or destroy any copy of this message and its attachments.

Mime
View raw message