The UDB Log Reader Capture actually performs two functions: Mining the UDB Log and Publishing captured data managed by the Storage agent. The Publisher pushes committed data directly to Engines using TCP/IP. The TCP/IP Publishing function manages the captured data until it has been transmitted and consumed by Connect CDC (SQData) Apply Engines, ensuring that captured data is not lost until the Engines, which may operate on other platforms, signal that data has been applied to their target datastores.