Mailing-List: contact dev-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hbase.apache.org
Received-SPF: pass (athena.apache.org: domain of yuzhihong@gmail.com
 designates 209.85.213.45 as permitted sender)
MIME-Version: 1.0
In-Reply-To: 
 <CACyo8+NWPXzW36HCU1mkP81MDM-TtSv5BzvX3LLP4L9gA_+MPw@mail.gmail.com>
References: 
 <CACyo8+NWPXzW36HCU1mkP81MDM-TtSv5BzvX3LLP4L9gA_+MPw@mail.gmail.com>
Date: Tue, 13 Jan 2015 14:16:23 -0800
Message-ID: 
 <CALte62zyK-rQcMHxWXCKUmFb4QCp5hYik=6B563597b1eQOweg@mail.gmail.com>
Subject: Re: regions per server
From: Ted Yu <yuzhihong@gmail.com>
To: "dev@hbase.apache.org" <dev@hbase.apache.org>
Content-Type: multipart/alternative; boundary=001a11c2be0c2862a3050c8ff91f

--001a11c2be0c2862a3050c8ff91f
Content-Type: text/plain; charset=UTF-8

Have you read http://hbase.apache.org/book.html#ops.capacity.regions.count ?

Cheers

On Tue, Jan 13, 2015 at 2:13 PM, Krishna <research800@gmail.com> wrote:

> Hi,
>
> Current HBase cluster has close to 800 regions per RS with data uniformly
> distributed across the nodes. could someone share their thoughts on
> following options for efficiently decreasing # of regions/server?
>
> * Increase region file size from 10GB to 50 GB: What is the impact on
> queries by increasing region size from 10GB to 50GB? Would query latency
> increase 5 times?
> * Add additional servers to the cluster: Why is [20-200] an ideal range for
> number of regions per RS? Can a RS handle over 1000 regions?
>
> Best...
>

--001a11c2be0c2862a3050c8ff91f--