|
Topics: |
Magnify Search requires an application server (for example, Tomcat or WebSphere) in order to read and write to index libraries. It can be installed on a Windows (64-bit preferred) or UNIX\Linux-based server, using the same language as the WebFOCUS Client.
Processing should be at least a dual-core machine. However, four CPUs are generally recommended. Moreover, a minimum of 8 GB in memory is required, while 16 GB is usually recommended. Depending on the size of the data, storage space should be anywhere between 250 GB to 1 TB or higher (it is usually estimated as three times the size of the expected dataset).
Note: This all varies depending on the type of search-based application being proposed. Indexing frequency, volume of data, and concurrent usage affect how storage space and memory are allocated. Therefore, the use of a split-tier environment is recommended for a Magnify Search platform where the Reporting Server, Magnify Search, and each data source are each provided with their own dedicated resources. Magnify Search does scale in order to accommodate any environment.
The following are the prerequisites to configuring and using Magnify Search:
You may need to adjust the Java VM memory options to correct performance issues, or after receiving a Java Heap Space error message.
Note: This is applicable at the application server level for any version of Magnify Search.
The most common Java VM options you need to set include the size of the Java heap and stack, which determine memory availability for Java programs and the Java VM. Errors can occur if not enough memory is available, and the heap size impacts performance, since it determines how often garbage collection occurs.
As a best practice, add the following setting to the JVM arguments when starting Apache Tomcat:
-XX:MaxPermSize=256m
Another setting that you may also need to increase is the general memory setting. For example:
-mx3000m
| WebFOCUS | |
|
Feedback |