3.5.1 Matrix Server Administration Guide

Chapter 1: Introduction 2

on a SAN. After a PSFS filesystem has been created on a SAN disk, all

servers in the matrix can mount the filesystem and subsequently

perform concurrent read and/or write operations to that filesystem.

PSFS is a journaling filesystem and provides online crash recovery.

• Availability and reliability. Servers and SAN components

(FibreChannel switches and RAID subsystems) can be added to a

matrix with minimal impact, as long as the operation is supported by

the underlying operating system. Matrix Server includes failover

mechanisms that enable the matrix to continue operations without

interruption when various types of failures occur. If network

communications fail between any or all servers in the matrix, Matrix

Server maintains the coherency and integrity of all shared data in the

matrix.

• Matrix-wide administration. The Management Console (a Java-based

graphical user interface) and the corresponding command-line

interface enable you to configure and manage the entire matrix either

remotely or from any server in the matrix.

• Failover support for network applications. Matrix Server uses virtual

hosts to provide highly available client access to mission-critical data

for Web, e-mail, file transfer, and other TCP/IP-based applications. If a

problem occurs with a network application, with the network

interface used by the virtual host, or with the underlying server,

Matrix Server automatically switches network traffic to another server

to provide continued service.

• Administrative event notification. When certain events occur in the

matrix, Matrix Server can send information about the events to the

system administrator via e-mail, a pager, the Management Console, or

another user-defined process.

MxFS-Linux provides the following features:

• Scalable NFS client connectivity. Over multiple NFS servers sharing

the same filesystems, MxFS-Linux supports a linearly increasing client

connection load as similarly configured servers are added to the

cluster. A 16-node cluster, serving the same filesystems via NFS, can

support 16 times more NFS clients (with similar workloads and the