gtpg1m0k | General Information |
Figure 11 shows communication between the following distinct TPF system communication configurations:
Figure 11. Communication Between Non-SNA and SNA Facilities
In a high-performance telecommunications system, the time required to carry out a restart after a system interruption must be short. To keep restart time to a minimum, the TPF system does not automatically reinitialize the network on restart. The host and cluster controllers preserve message sequence numbers over a system interruption, and no network initialization is required. Error conditions such as NCP slowdown, automatic network shutdown, and lost input or output messages are handled as exception conditions by error recovery routines.
Support of inbound and outbound message traffic pacing between application programs or supported terminals helps avoid buffer overrun.
Dynamic collection of tuning statistics support is also provided. I/O interface data can be dynamically collected. Such data can aid a user in selecting the best values for system parameters.
The TPF system fully supports the message integrity capability provided in SNA by message sequence numbering. As a system generation option, the TPF system provides a message recovery package to ensure that each input message is processed successfully and that its reply is delivered successfully. The level of recovery is user-selected, and it extends from full recovery of all input and output messages to simply keeping track of each input message being processed. The TPF system will maintain the message security over a system interruption. Message security and integrity facilities differ according to line protocol and remote devices.
The TPF system also provides an application-oriented recovery package that allows message recovery and database recovery to be combined. In addition, the TPF system provides support to encrypt and decrypt messages to or from any device.
The primary responsibility for control of the SNA network lies with the VTAM CMC. The TPF operator has the ability to control and monitor TPF system attachment to the network through operator commands. The status of an NCP, a cross-domain resource manager (CDRM), or an individual logical unit can be displayed at the operator console.