That would be my recommendation as well:  TCP window size is the most likely culprit.

 

Windows (newer versions) will negotiate a very large window size which can help.  The filer’s default has been 17424 for some time, which will make things slower.

 

As for the SMB vs CIFS thing, they’re essentially the same.  If you’re using Windows, you’re using CIFS (unless you’ve loaded services for unix, but I’m pretty sure you’d realize if you were utilizing NFS by accident).

 

Glenn

 


From: Shane Garoutte [mailto:sgaroutte@gmail.com]
Sent: Wednesday, March 28, 2007 3:29 PM
To: Langdon, Laughlin T. ((Lock))
Cc: Glenn Walker; toasters@mathworks.com
Subject: Re: CIFS overhead with Netapp Filers

 

A quick crawl on NOW provided the following:

http://now.netapp.com/Knowledgebase/solutionarea.asp?id=ntapcs675

 

if CIFS performance is slow after investigating performance issues, modify the filer's CIFS negotiated buffer size.

 

   1. Verify that hardware or software problems do not exist within the filer, network and client.

   2. Record the CIFS negotiated buffer size by capturing the output of the filer command:

      options cifs.neg_buf_size

   3. Enter the following filer commands:

      a. cifs terminate

      b. options cifs.neg_buf_size 16644

      c. cifs restart

   4. If the buffer size in step 3b does not improve performance, try the following buffer sizes:

      a. Use '17424'.

      Note:

      Starting with Data ONTAP 6.0.X, allow the buffer size to exceed 17424; therefore, upgrade to a release that fixes bug 33396 only if performance does not improve.

      b. Use '33472' for environments mixed with Windows NT and Windows 2000.

      c. Use '65340' for Windows 2000 only environments.

   5. If performance remains slow:

      a. Re-confirm that hardware or software problems do not exist within the filer, network and client.

      b. Restore the original CIFS negotiated buffer size (refer to steps 2 and 3).

      c. During a performance interruption, capture a packet trace between the filer and Windows client.

      d. Send the packet trace to Network Appliance Technical Support for analysis.

 

 

On Mar 28, 2007, at 8:33 AM, Langdon, Laughlin T. ((Lock)) wrote:



I’m doing a straight drag and drop using UNC paths with  a single 1.5gig zip file and a 2.2Gig binary File.  If I add more streams (aka start more than one copy on more than one server the filer happily provides more bandwidth)

 

From Windows server to windows server I get 500 Mbps

From Windows server to a Netapp 6030 Filer running DOT 7.2.1 I get about 250 Mbps

 

I’ve tried TCP windows size, Flow Control, LCAP, Static Link Aggregation, Singe port on the filer (no vif), straight crossover cable.

 

Glenn Walker [mailto:ggwalker@mindspring.com]
Sent: Tuesday, March 27, 2007 5:15 PM
To: Langdon, Laughlin T. (Lock); toasters@mathworks.com
Subject: RE: CIFS overhead with Netapp Filers

 

 

 

 

 

size=2 width="100%" align=center>

owner-toasters@mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Langdon, Laughlin T. (Lock)
Sent: Tuesday, March 27, 2007 2:33 PM
To: toasters@mathworks.com
Subject: CIFS overhead with Netapp Filers