Application Discovery Service (closed to new customers)
Application Discovery Service collected configuration and usage data from on-premises servers so that a migration could be planned against evidence rather than against a spreadsheet somebody maintained by hand. It fed Migration Hub, where discovered servers were grouped into applications and tracked through the move.
The two collection modes
Section titled “The two collection modes”Both modes still describe how discovery works in general, whichever tool performs it.
Agentless Collector — an OVA appliance deployed into VMware vCenter. It enumerates hosts and virtual machines and reports IP and MAC addresses, resource allocation, host names and utilisation metrics. Nothing is installed on the guests, so it is the mode that gets approved fastest in a change-controlled estate, and the mode that reveals least about what is running inside each machine.
Discovery Agent — a small agent installed on each virtual or physical server, with installers for Windows and Linux. It reports static configuration, time-series performance data, network connections and running OS processes. The network connections are the point: they are what turns a list of servers into a dependency graph, which is what wave planning needs.
What replaces it
Section titled “What replaces it”AWS Transform runs both agent-based and agentless collection, adds storage and database discovery, consolidates third-party inventory tools, and produces dependency maps and prioritised migration waves directly rather than handing raw data to a separate planning step. Moving across requires no data migration; new discovery projects simply start there.