How Does SNA over TCP/IP Work?

SNA over TCP/IP uses protocols that bypass the lower transport layers of the SNA architecture. This process is transparent to the SNA application program. Using a supported API, an application program passes data to the presentation services of Personal Communications. This data, in turn, is passed through the SNA architectural layers and presented to SNA over TCP/IP at the data link control level. When Personal Communications initiates a session for an application program, SNA over TCP/IP translates the SNA routing information (network-qualified name) into IP routing information (IP address) and uses the IP address to create a TCP connection to the appropriate system.

Figure 1 illustrates how an SNA over TCP/IP access node enables SNA application programs to communicate over IP networks.

Figure 1. Structure of SNA over TCP/IP for Personal Communications

The configuration information you define enables SNA over TCP/IP to determine:

Whether to route the data using SNA or to route the data using IP
The IP address associated with the network-qualified LU name

SNA over TCP/IP uses both stream (TCP) and datagram (UDP) sockets that are bound to the well-known port (397). Any information received over this port is routed to AnyNet®.