Updated June 16, 2016
This article provides brief information on how UDT gathers the hostnames through DNS for the endpoints it discovers off your Layer 2 and Layer 3 devices.
All UDT versions
UDT gathers its hostname information from whichever DNS server the UDT host server is configured for. It gathers this based off any PTR records present on the server affiliated to the IP queried. How it reaches this point though is the last part in a chain of polls conducted to get all the endpoint information.
UDT first starts by determining the MAC address information for the endpoints attached to your switch devices in UDT. Once that information is gathered through it's Layer 2 polling, UDT then conducts a Layer 3 polling against that node (if configured for it). This is to gather the ARP data from that device, where it then upon compares the MAC in the ARP results to the MAC in the endpoint listings found off the switches. If it matches up, then it assigns that IP from the ARP to the MAC.
At the end of the Layer 3 poll, UDT then queries for the hostname of the IP based on whether it is a new or existing, current endpoint entry. If it's a new IP, then it immediately queries the DNS server for the name. If it is an existing IP already though, it handles the query a bit different.
UDT stores any IP that is current in the database in a system cache for later querying, depending on whether it succeeded or failed before. This is known as the Positive and Negative DNS Cache.
These Cache timing values are based per IP, not in bulk. This means that each IP will have it's own cache results so UDT avoids doing mass DNS queries to the DNS server all at once. These values can be changed per your preference.