You can streamline the process of registering agents on clusters by defining a DNS name for a cluster and configuring that cluster so that the metrics are shared sequentially in a loop.
You only need to register the agent on the DNS, not on the IP address of each individual machine in the cluster. If you do register the agent on each node in the cluster, it affects the scale of your environment.
When you have configured the cluster so that the received metrics are shared in a sequential loop, each time that the agent queries the DNS server for an IP address, the returned address is for one of the virtual machines in the cluster. The next time the agent queries the DNS, it sequentially supplies the IP address of the next virtual machine in the cluster, and so on. The clustered machines are set up in a loop configuration so that each machine receives metrics in turn, ensuring a balanced load.
After you configure the DNS, it is important to maintain it, ensuring that when machines are added or removed from the cluster, their IP address information is updated accordingly.