What is an xP3 SCpbMD deployment?

The xP3 SCpbMD deployment is set up for environments with higher load. All roles have been broken out onto their own servers, leaving the CD server with more resources available for serving content. Having all roles on separate servers means that if any servers experiences an increase in load, you can easily scale horizontally or vertically. An xM3 deployment would be identical to this except there would be no xDB.

This deployment also includes a separate search role, to be used with a search product such as Solr or Coveo, which is hosted on a different server. It is strongly recommended that you use Solr in all production scenarios, and use Lucene for development only. Using Lucene only for development and having a separate search role means you no longer need to worry about indexes needing to be copied to each CD server, or resources on each CD server and the disk/CPU becoming completely consumed during indexing.

With higher traffic sites, xDB may have a high level of engagement activity that needs to be condensed into the Reporting database. This processing is performed by the Processing role and uses a large amount of CPU resources, so you may want to place the Processing role on a separate server.

With this topology, the Routing Service must be put on the same server as the Content Management server, but the Transaction service can be put on its own server. If you have a high level of traffic, you will want to leave the Transaction service on its own server so that the CD servers can focus solely on serving content, and you can dedicate a full set of CPUs to work with Dynamics AX.