Show TOC

Distributed TREX Systems (Multiple Host Installation)Locate this document in the navigation structure

Use

Search and Classification (TREX) consists of a client component and a server component. The server component is based on a flexible architecture that allows a distributed installation. A distributed system has the following advantages:

  • Load balancing

    You can distribute the search and indexing load among several hosts.

  • High Availability

    You can make searching and indexing highly available.

This guide explains how to plan and implement a distributed system. It is aimed at technology consultants.

The guide is structured as follows:

  • Naming Conventions contains information on the naming conventions used in this guide.

  • Required Documentation lists the documentation that you need to implement a distributed system.

  • The TREX Fundamentals section

    contains information about the TREX architecture and basic information about distributed systems. You need this information to plan a distributed system. Read this information before you begin to implement your distributed system.

  • The sections Setting Up a Distributed System

    and Delta Index and Index Replication Configuration

    explain how to implement a distributed system.

  • The section Changing a Distributed System

    describes the changes that you can make to a distributed system after the installation.

  • The Distributed Preprocessing of Documents section

    describes how you can distribute document preprocessing among more than one host. This section is relevant if you want to index documents whose preprocessing takes up a lot of time and system resources. This can be the case if you want to index large PDF files.

  • The appendix contains information on stopping and starting a distributed system. It also contains information on starting the TREX admin tool and changing the queue parameters and the Java client parameters.

Note

Restricted support for the global file system function for TREX 7.1.

TREX 7.0 allows you to install a central TREX instance on a global file system. On a global file system, you can store files for the SAP profiles centrally so that other TREX dialog instances can access them. You can use this function to set up distributed TREX landscapes. TREX 7.1 currently does not support the global file system function.

In the TREX 7.1 installation, the global file system function is available in restricted form only. The customer can no longer specify the path to the global file system during the installation process (SAP system directory usr/sap/<sid>/sys). The global file system is installed once only, on the host on which you initially install TREX 7.1 and it cannot be installed again independent of this. Therefore, there is still a global file system in the TREX 7.1 installation, but it cannot be installed on another host in the TREX landscape independent of the initial TREX 7.1 installation. TREX is developing the unrestricted global file system feature in TREX 7.1 for the end of 2008.