Chapter 16: The Network File System


Overview

The Network File System (NFS) looks like a single, unified filesystem to users and application programs running on the cluster nodes. A key feature of NFS is its ability to share data with all of the cluster nodes while providing a locking mechanism to protect the integrity of the data. To build a cluster that does not require any changes to existing multiuser application programs, this locking mechanism must support the locking methods used by programs that are unaware of the fact that they are running on a cluster node. That is, NFS and its locking mechanism should be transparent to the existing multiuser applications. This transparency of the shared filesystem is one of the defining characteristics of the Linux Enterprise Cluster—applications see the cluster as a single, unified computing resource.

In this chapter, we'll examine what the term lock means, what locking methods are normally used on multiuser operating systems, and how NFS allows existing legacy, multiuser applications that are running on cluster nodes to share access to data. (If you do not need to run legacy Unix applications on your Linux Enterprise Cluster, you can skip over most of the locking information in this chapter.) I'll also introduce you to a few performance issues you will face when using NFS and provide you with an example of the NFS client configuration options you can use on each node inside the cluster.

Note 

In Chapter 20, we'll look at how a database system such as MySQL or Postgres can be used in a cluster environment.



The Linux Enterprise Cluster. Build a Highly Available Cluster with Commodity Hardware and Free Software
Linux Enterprise Cluster: Build a Highly Available Cluster with Commodity Hardware and Free Software
ISBN: 1593270364
EAN: 2147483647
Year: 2003
Pages: 219
Authors: Karl Kopper

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net