Boston Linux & Unix (BLU) Home | Calendar | Mail Lists | List Archives | Desktop SIG | Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings
Linux Cafe | Meeting Notes | Blog | Linux Links | Bling | About BLU

BLU Discuss list archive


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

NFS performance question



 I have 6 Woodcrest machines with one acting as an NFS server. 
3 machines are in the server room plugged directly into a Netgear 1GB 
switch. These 3 are all running RHEL4 U 3. The other 3 machines are in 
a rack  in our office. They are all connected to a Netgear 1GB switch 
that is connected to a wall jack that goes into a patch panel in the 
server room and eventually ends up at the initial switch.  The 3 
systems in the rack are RHEL4U6. 

When we run our application, then save the data on both the NFS server 
and one of the NFS client machines in the server room, the save takes 
just a second or 2. But, when I save on one of the servers in the 
office the save takes nearly a minute.  I checked the 2 switches and 
they show all relevant connections as being 1GB. So, the differences 
between machines are: 
1. NFS client in server room RHEL4U3, connected directly to the switch. 
2. NFS client in office RHEL4U6 connected to a 1GB switch that is 
connected via an office patch panel to the switch in the server room. 

Note that when I ping the systems from the server room, I get similar 
responses (under .5ms) All the systems are on the same subnet. 

I am not seeing any nfs related messages on any of the client machines. 

Eventually all the servers will be in the rack on the same switch, but 
I don't want to put the rack into the server room until we get our KVM. 
My thoughts are the delay may have something to do with latency. 

I ran a network test, thrulay, between several of the NFS client 
servers and the NFS server, all with pretty decent bandwidth: 
936.5 Mbps and 940.6 Mbps on the servers in the office, and 940.7 and 
941.3 on the servers in the server room. The 2 office servers were 
being used more heavily at the time, but I would say the data rates are 
consistent. 


I've got plenty of memory on those systems in the server room (16GB) 
and I'm not showing any swap activity. 

The question is any idea why the 2 RHEL4U6 systems seem to have such 
poor NFS performance. 

-- 
-- 
Jerry Feldman <[hidden email]> 
Boston Linux and Unix 
PGP key id: 537C5846 
PGP Key fingerprint: 3D1B 8377 A3C0 A5F2 ECBB  CA3B 4607 4319 537C 5846 
_______________________________________________ 
Discuss mailing list 
[hidden email] 
http://lists.blu.org/mailman/listinfo/discuss
 


BLU is a member of BostonUserGroups
BLU is a member of BostonUserGroups
We also thank MIT for the use of their facilities.

Valid HTML 4.01! Valid CSS!



Boston Linux & Unix / webmaster@blu.org