AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:<20080514094422.499504ce@ripper.onstor.net>
CFG:
PT:0
S:andy.sharp@onstor.com
RQ:
SSV:onstor-exch02.onstor.net
NSV:
SSH:
R:<jhernandez@css.glasshouse.com>,<dl-cstech@onstor.com>
MAID:1
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/andys@onstor.net@onstor-exch02.onstor.net/INBOX	0	200805141630.m4EGUeg24973@mailhost-rtp.css.glasshouse.com
X-Sylpheed-End-Special-Headers: 1
Date: Wed, 14 May 2008 09:44:35 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: "Jairo Hernandez" <jhernandez@css.glasshouse.com>
Cc: "'dl-cstech'" <dl-cstech@onstor.com>
Subject: Re: case 8168--[Ask.com--Onstor handling smaller files]
Message-ID: <20080514094435.3a96acb0@ripper.onstor.net>
In-Reply-To: <200805141630.m4EGUeg24973@mailhost-rtp.css.glasshouse.com>
References: <200805141630.m4EGUeg24973@mailhost-rtp.css.glasshouse.com>
Organization: Onstor
X-Mailer: Sylpheed-Claws 2.6.0 (GTK+ 2.8.20; x86_64-pc-linux-gnu)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

On Wed, 14 May 2008 11:30:48 -0500 "Jairo Hernandez"
<jhernandez@css.glasshouse.com> wrote:

> Hello guys,
> 
>  
> 
> I have Jay Patel from Ask.com asking the following. Hopefully someone
> can provide feedback.
> 
>  
> 
> *******************************************************************
> 
>  
> 
> We ran some test on our netapps systems and found : 
> 
> 1. The files on the WAFL filesystem makes files <=64 bytes don\'t
> have any blocks allocated to them , it fits data into the inode
> entry. A really old filesystem called ext2 also handles files <=64
> bytes the same way. 

No, ext2 does not.  But that's really quite irrelevant.

> For the same command 
> 
> find 470/948/littlemissindie -printf \"%y %p %s %b\\n\ <file:///\\n\>
> " 
> 
> Netapps shows 
> y 470/948/littlemissindie/friends--sf1.-1.0.2/16b/27/nibs007 52 0 <<<
> 0 blocks???? 
> y 470/948/littlemissindie/friends--sf1.-1.0.2/191 4096 8 
> 
> OnStor shows 
> f 470/948/littlemissindie/friends--sf1.-1.0.2/16b/27/nibs007 52 16
> <<< d 470/948/littlemissindie/friends--sf1.-1.0.2/191 1024 16 
> 
> 
> Why doesn\'t ONstor do it the same way and how does ONSTOR write
> it\'s files for smaller files? 

I do believe that we put very small files in the inode, but I'll let
the Joberator be the final authority on that.

> What should we do to handle the extremely large storage requirements
> as per this issue and also the fact that the 8k default block size
> also increase storage requirements by a big factor? 
> 
> Jay       
> 
> *******************************************************************
> 
>  
> 
> Thanks,
> 
>  
> 
> jairo
> 
