PC Server 720 Information and Software
System Type: 8642

Features Unique to the Micro Channel Bus

The Micro Channel bus provides features not implemented in the PCI bus. These include the streaming-data procedure, channel-check reporting, and error logging.

Streaming-Data Procedure

The streaming-data procedure allows high-speed data transfers between bus masters and subordinates. This procedure supports high-speed transfers of large blocks of data for devices such as hard disk drives and network adapters. The streaming-data procedure transfers blocks of sequentially stored data. In basic data-transfer operations, a target address is assigned for every byte of data that is transferred. The streaming-data procedure assigns a target address only to the first byte of data; the rest of the data in the block follows in sequence and is assigned to a sequential address. The streaming-data procedure supports a data-transfer rate of 80 MBps, which is four times as fast as basic data-transfer operations. The Micro Channel bus has 32 data lines and 32 address lines. During streaming-data transfer operations, the 32 address lines are used only at the beginning of a transfer cycle; then, they remain idle for the rest of the cycle. The multiplexed streaming-data procedure uses the address bus as another 32-bit data bus, allowing data to be transferred 64 bits at a time.

Channel-Check Reporting and Error Logging

Errors can occur not only during data-transfer operations, but also while data is stored in system memory. The contents of a memory location can be changed accidentally, a memory module can be defective, or other hardware failures can occur in the server. Your server uses channel-check reporting to detect hardware errors and error logging to record the errors. Service technicians can use these records to diagnose and correct problems in the server. The channel-check reporting facility automatically locates random and intermittent errors while your server is operating. Information about any failing component is saved in the error log so that you can identify and replace the failing component.