5 Architectural Considerations for Using Autonomy Search

During the architecture phase, you determine what enterprise content you want to make available for your portal and who within your portal environment will be able to search this information.

This chapter includes the following sections:

Section 5.1, "Understanding How Autonomy Search is Implemented"
Section 5.2, "Deciding What Information to Index"
Section 5.3, "Architectural Recommendations"

5.1 Understanding How Autonomy Search is Implemented

Oracle WebLogic Portal uses Autonomy search components to implement search functions such as allowing portal users to search external web sites, integrated databases, and available file systems. To do this, you can incorporate search portlets that ship with Oracle WebLogic Portal or write your own. However you decide to implement search, the same tools will be used. You will need to install and configure these tools before integrating search within your portal.

Note:

If you upgraded to Oracle WebLogic Portal 10.3.4, search portlets are included in your installation. If you are performing a clean install of Oracle WebLogic Portal and want to use Autonomy's search portlets, you must obtain them from Autonomy Corporation.

The Autonomy IDOL server is used to manage the indexes created by the Autonomy fetches you use. You can use this indexed data in your portal application by using the portlets that come with Oracle WebLogic Portal or you can write your own portlets using the Autonomy API. For more information about using and developing portlets, see Chapter 3, "Configuring Your Portal to Use Autonomy Search."

Figure 5-1 shows a diagram of how Autonomy search tools integrate with Oracle WebLogic Portal.

Figure 5-1 Diagram of an Oracle WebLogic Portal Integration with Autonomy Search Tools

Description of "Figure 5-1 Diagram of an Oracle WebLogic Portal Integration with Autonomy Search Tools"

5.2 Deciding What Information to Index

You can allow portal users to search a variety of information sources from your portal. Before developing and deploying your portal, you should decide what information sources you want to index. Consult the respective Autonomy documentation for more details. For example, if you want to include web sites, see the Autonomy HTTP Fetch Administrator's Guide, published by Autonomy Corporation. Contact WebLogic Portal Customer Support to obtain a copy of this guide.

5.2.1 Maximum Amount of Indexed Content

If you upgraded to Oracle WebLogic Portal 10.3.4, the Autonomy license that is delivered with Oracle WebLogic Portal allows you to index 500,000 pieces of content. A single IDOL server can hold an approximate maximum of 8 million document sections (depending on the functionality and performance required). However, the license included with previous versions of Oracle WebLogic Portal allows only 500,000. The current version of Oracle WebLogic Portal does not include an Autonomy license.

If you performed a clean installation of Oracle WebLogic Portal and purchased Autonomy from Autonomy Corporation, the maximum amount of indexed content available is determined by your license agreement with Autonomy Corporation.

5.3 Architectural Recommendations

Oracle recommends using a separate machine for your Autonomy IDOL Server to ensure the most processing power to service indexing and query requests from your portal clients. You can also install each Autonomy engine (such as HTTP Fetch, File System Fetch and IDOL Server) on a separate server if you find you need additional resources.

Autonomy recommends a dual-processor server for hosting the IDOL Server and the DiSH Handler. For complete system requirements, see the Autonomy documentation.