IBM Support

Better performance with NFS server and client

Question & Answer


Question

How do I tune NFS when using RHEL 5.3 on an IBM Netezza 1000 system?

Answer

Configuring the rsize and wsize in the /etc/fstab file for an NFS mount point can change the data transfer speed between an NFS server and NFS client. To achieve better performance in RHEL 5.3, edit /etc/fstab and remount the volume.

The rsize and wsize are typically set at between 16k and 128k for best performance. The "best" value is heavily dependent on your NFS server. For more information on optimizing NFS performance, refer to sourceforge.net recommendations.

Note: The instructions below assume that you have an existing NFS mount configured. Substitute your IP address and mount points, based on your configuration, in the examples below.

1. Edit the /etc/fstab file by adding the following line:


# vi /etc/fstab
192.168.1.200:/tmp      /export/home/nfstest  nfs    defaults,nolock,rsize=32768,wsize=32768,intr,nfsvers=3,soft
2. Unmount and then remount the volume:
# umount /export/home/nfstest
# mount /export/home/nfstest

3.  Verify that the settings are correct using the mount command, as follows:
[root@cricket ~]# mount | grep nfs 
sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)  
192.168.1.200:/tmp on /export/home/nfstest type nfs (rw,noatime,nolock,rsize=32768,wsize=32768,intr,nfsvers=3,soft,addr=192.168.1.200)
[root@cricket ~]#
[{"Product":{"code":"SSULQD","label":"IBM PureData System"},"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Component":null,"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"1.0.0","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Historical Number

NZ160575

Document Information

More support for:
IBM PureData System

Software version:
1.0.0

Document number:
458109

Modified date:
17 October 2019

UID

swg21564031

Page Feedback