We are in the process of installing and configuring SCOM 2016 with 4 VMs for management server, 1 for Database and 1 for Datawarehouse and reporting server. Expecting to go with agentless exception monitoring for desktop and laptops with windows 10. Would like your help in how these 4 VM's adm file can be imported in Active directory group policy for 40000 and above clients to equally communicate with all management servers to avoid proper memory and CPU utilisation. Also is there any tool for agentless exception monitoring to check how many clients are communicating to one management server VM. As per the document 100000 clients can communicate in agentless exception monitoring on one management server VM, so can you please suggest some tweaks to make this work/ function correctly without making management server go critical or greyed out.