Archives


- Beowulf
- Beowulf Announce
- Scyld-users
- Beowulf on Debian

NFS 100:1 performance loss

Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.

Search

Martin Siegert siegert at sfu.ca
Fri Aug 18 15:03:10 PDT 2000


On Fri, 18 Aug 2000, Josip Loncaric wrote:

> Has anyone seen this NFS performance problem?
> 
> Our Beowulf has two servers w/RAID-0 arrays which deliver 45MB/s,
> connected via Gigabit Ethernet.  The machines are dual PIII/500 systems
> with 512MB of RAM each.  We are running Red Hat 6.2 updated to kernel
> 2.2.16-3 and recently we updated to nfs-utils-0.1.9.1-1.  We are running
> 16 kernel nfsd threads on each machine.  Here is how long it takes to
> copy a 28,955,860 byte file from machine 1 to machine 2:
> 
> rcp:       1.04 seconds (27.8 MB/s, where 1MB=10^6B)
> ftp:       1.12 seconds (25.8 MB/s)
> NFS 1KB:  12.35 seconds ( 2.3 MB/s, using rsize=wsize=1024)
> NFS 8KB: 129.42 seconds ( 0.2 MB/s, using rsize=wsize=8192)
> 
> Clearly, there is something very wrong with NFS, particularly with 8KB
> rsize/wsize (which should have improved performance!).  Our system
> manager tells me that reduced Linux NFS performance with 8KB rsize/wsize
> is a known problem, but even at 1KB our NFS is getting less than 10% of
> the rcp or ftp performance.
> 
> Any ideas?

No ideas. Sorry. But here are the numbers from our system:
dual PII/400, 256MB RAM
kernel-2.2.16-3
nfs-utils-0.1.9.1-1
No RAID, etc., just IDE drives, /home and /usr/local NFS exported from the
master node.

Contents of /etc/fstab:
...
b01:/usr/local          /usr/local              nfs     ro,hard,intr,bg,rsize=8192 0 0
b01:/home               /home                   nfs     rw,hard,intr,bg,rsize=8192,wsize=8192 0 0

# cd /usr/local/src/tar-files
# ls -l linux-2.2.16bugaboo.tar
-rw-r--r--   1 root     root     81674240 Jun 28 15:13 linux-2.2.16bugaboo.tar

# time cp linux-2.2.16bugaboo.tar /tmp

real    0m9.799s
user    0m0.040s
sys     0m2.150s

That's about 63.59 Mbit/s.
Second test (after "gzip linux-2.2.16bugaboo.tar" on master node):
# ls -l linux-2.2.16bugaboo.tar.gz
-rw-r--r--   1 root     root     20673033 Jun 28 15:13 linux-2.2.16bugaboo.tar.gz
# time cp linux-2.2.16bugaboo.tar.gz /tmp

real    0m2.383s
user    0m0.000s
sys     0m0.280s
That's about 66.19 Mbit/s.

In the other direction:
# cd /tmp
# time cp linux-2.2.16bugaboo.tar.gz /home/siegert

real    0m1.854s
user    0m0.000s
sys     0m0.270s
That's about 85.07 Mbit/s.

This is with 100baseT and 3C905B NICs.
So I doubt that it is NFS that is at fault here ...

Cheers,
Martin

========================================================================
Martin Siegert
Academic Computing Services                        phone: (604) 291-4691
Simon Fraser University                            fax:   (604) 291-4242
Burnaby, British Columbia                          email: siegert at sfu.ca
Canada  V5A 1S6
========================================================================





More information about the Beowulf mailing list