Configure a processing server¶
You can configure one or multiple processing servers for aggregation or other kinds of processing. Multiple processing servers can provide better scalability and performance. As a general rule, the more aggregation server instances you create, the faster aggregation processing will be.
- If you run the xDB Cloud edition, then configure your processing environment according to the instructions in this topic and the instructions for configuring xDB Cloud.
You should synchronize all servers to a single reliable time source, such as with the Network Time Protocol (NTP). Engagement automation state aggregation depends on the system time. Inconsistency in time sources can lead to incorrect aggregation results or loss of data.
Before you install a Sitecore server you must read the hardware guidelines for Sitecore Hosting Environment Requirements and Sitecore Client Requirements sections from Installation Guide 9.0. You can download the guide from the Sitecore Downloads page.
You can install a Sitecore server either manually or by using the Sitecore Installation Framework. We recommend you use the Sitecore Installation Framework according to the instructions in the Installation Guide 9.0. The Sitecore Installation Framework automatically configures your Sitecore server for the role you select.
If you install the server manually, you must configure it for the processing role:
Use rule-based configuration to configure the server to fulfill only the processing role by including this line in the web.config file:
<add key="role:define" value="Processing" />
Ensure the following connection strings are set in the \App_Config\ConnectionStrings.config file:
Name Type Notes core SQL master SQL reporting SQL Collection data is aggregated into the reporting database. xconnect.collection HTTPS xconnect.collection.certificate Certificate xdb.referencedata SQL Accesses definition data during processing. Processing is an xDB component, and accesses the database directly. xdb.processing.pools SQL reporting.apikey Unique string Required if you use a dedicated Reporting Service or processing server. xdb.processing.tasks SQL
If you have set up xConnect to use separate collection and search end points, you must configure them separately.
Configure content search. Contact and interaction search is done through the xConnect Search service, as CM servers do not have direct access to the xDB core. Configuration depends on your provider:
- Configure Solr
- Configure Lucene
In a scaled environment, you should use Solr or Azure Search. Lucene is only recommended for standalone servers. The Azure Search provider only supports Azure Cloud PAAS deployments.
Please submit documentation feedback to firstname.lastname@example.org.