Upgrade summary

The upgrade summary provides information about the components that are supported in the Watson Content Analytics Version 3.5 system and summarizes the supported migration paths.

For items that include data that can be migrated, the table shows whether the migration is handled automatically by the installation program or whether manual steps are required to finalize the upgrade.

Table 1. Upgrade Summary. Summary of supported migration paths
Functional Area	Automatic	Version 3.5 Requirements
Topology
Single server installation		If you currently run the product on a single server configuration, you must upgrade to a single server configuration. You can then add servers to support search, document processing, and indexing.
Distributed server installation		If you currently run the product on a distributed server configuration, you must upgrade to a distributed server configuration. You can then add servers to support search, document processing, and indexing.
High availability servers		Supported on AIX® and Windows systems.
IBM® InfoSphere® BigInsights server		Supported on Linux 64-bit systems.
Configuration
Crawlers	Automatic	Crawler configurations are automatically migrated to Version 3.5. If you prefer, you can manually import crawler configuration data to Version 3.5.
Backup and Restore		You cannot create a backup of an older version of the product and restore it on a Version 3.5 system.
Importing and exporting collections between different versions		You cannot import and export collections between different versions of the product.
Document cache	Automatic	Cached data is migrated and does not need to be recrawled. To apply new features to previously crawled data, however, the data must be recrawled.
In process data		Documents that were crawled but not yet parsed are not migrated. Use the administration console to ensure that no documents are waiting to be parsed before you upgrade to Version 3.5.
Administration	Automatic	Extended administration configurations can be imported to Version 3.5. For example, you might have `component`_ext.xml files to modify default crawler behavior and other administration functions. Examples of ways that you can extend configurations include: Increasing the maximum size of documents crawled by file system crawlers Configuring multiple threads for file system crawlers Configuring an isolation level for crawling DB2® databases Customizing how Lotus Notes® documents are recrawled Indexing document-level security for Lotus Notes documents to improve search performance Configuring a Windows file system crawler to support trusted domains Creating a plug-in to extract entries from archive files so that individual files can be parsed
Index
Search index	Automatic	The index is migrated and does not need to be rebuilt. To apply new features to previously indexed data, however, you must rebuild the index.
Analytics index	Manual	You must redeploy all analytic resources and rebuild the analytic indexes after you upgrade to Version 3.5.
Index on IBM InfoSphere BigInsights server	Automatic	The index is migrated and does not need to be rebuilt. To apply new features to previously indexed data, however, you must rebuild the index.
Application Server
Embedded application server	Automatic	If you used the embedded web application server in Version 3.0, the upgrade installation program installs all web applications (ESSearchServer, administration console, enterprise search application, and content analytics miner) on the embedded web application server. After you upgrade to Version 3.5, you must open the Security view in the administration console and select your LDAP server type. You must also reconfigure any LTPA token parameters that you configured for your Version 3.0 system. LTPA data is not migrated to Version 3.5.
WebSphere® Application Server and WebSphere Application Server Network Deployment, non-cluster installation	Automatic	If you used WebSphere Application Server in Version 3.0, the upgrade installation program installs the web applications in the same WebSphere Application Server.
Application Programming Interfaces
SIAPI APIs		Support is provided for the Search APIs only. For custom administration applications, use the REST APIs.
Enterprise search application	Automatic	Configuration changes made through the Search Customizer are migrated.
Content analytics miner	Automatic	Configuration changes made through the Analytics Customizer are migrated.
Custom enterprise search application	Manual	You must back up and redeploy your custom applications.
Search portlet		Not all of the Version 3.5 enterprise search application enhancements are available in the Version 3.5 search portlet.
Custom search portlet	Manual	You must back up and redeploy your custom search portlets.
Plug-ins	Manual	You must back up and redeploy your custom plug-ins, such as custom plug-ins for crawlers, post-filtering results, exporting documents, and adding custom views to the content analytics miner.
Annotators
Provided annotators and dictionaries	Automatic	Different annotators and dictionaries are provided for enterprise search collections and content analytics collections.
Custom annotators	Manual	You must back up and redeploy your custom annotators. For a content analytics collection: If the document cache is enabled, redeploy the analytic resources and rebuild a full index. If the document cache is not enabled, redeploy the analytic resources and run a full crawl.