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 jdcryans@gmail.com designates
 209.85.210.169 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:sender:in-reply-to:references:date
         :x-google-sender-auth:message-id:subject:from:to:content-type;
        b=lkBvhR4Q1ux+u4jpSc3419WgkjIhi8jJH+7D9AyPwKQ6GsF3K6gfhc/jlW8LzrI3Ru
         lBDgIUf3vM9ZtFc3nPEMj8WN8iBAZGC8SkceCFW4ZPhWuRvolM3Icxm0ogJ/TeBUy2tK
         CVBA3LWC6n6/sRby+pQQvItGjuyEExIcPgzQk=
MIME-Version: 1.0
Sender: jdcryans@gmail.com
In-Reply-To: <AANLkTim6oD4zN2Ed5G0PR-Y=9XVYEJmOQLcu4JjjKAwi@mail.gmail.com>
References: <AANLkTim6oD4zN2Ed5G0PR-Y=9XVYEJmOQLcu4JjjKAwi@mail.gmail.com>
Date: Mon, 28 Mar 2011 09:23:56 -0700
Message-ID: <AANLkTi=WCmJCYERds9fb+2psLw6WKntKkDxTOCKHSU4V@mail.gmail.com>
Subject: Re: disk seek in range search
From: Jean-Daniel Cryans <jdcryans@apache.org>
To: dev@hbase.apache.org
Content-Type: text/plain; charset=ISO-8859-1

I think you are asking for a bit too much :)

Let's do it the other way, show us what you think are the answers to
your questions based on currently available documentation and by
looking at the source code, then I'm pretty sure someone will be happy
to verify it.

Start by looking at the bigtable paper, then use Lars George's blog posts like:

http://www.larsgeorge.com/2009/10/hbase-architecture-101-storage.html
http://www.larsgeorge.com/2010/01/hbase-architecture-101-write-ahead-log.html
http://www.larsgeorge.com/2010/05/hbase-file-locality-in-hdfs.html

And then try diving into the code.

Good luck!

J-D

On Sat, Mar 26, 2011 at 8:57 AM, Weishung Chung <weishung@gmail.com> wrote:
> Dear fellow HBase developers,
>
> Could someone educate me and let me know how to figure out the number of
> disk seeks involved in a range search (startRow to endRow specified in
> Scan). Also, could anyone give me the details of all the steps involved once
> the Scan for range retrieval is called? I know somehow it needs to figure
> out the regionservers used in hosting the rows but I still don't have a
> clear understanding the whole steps involved :( :( Also, there is a data
> index block in HFile, I was wondering how the index block is utilized in
> figuring out the location of all the rows.
>
> Thank you so much for satisfying my curiosity :)
>
> Have a good weekend and enjoy :)
>
> Wei Shung
>