Introduction cont #

Parallel systems #

Intrachip transfer is much faster than interchip transfer.

Distributed systems #

Real-time and embedded systems #

Other systems #

Organization #

• PC = program counter, the address of the next instruction
• IR = instruction register, the address of the current instruction
• MAR = memory address register, address of the next memory IO
• MBR = memory buffer register, actual data to be read/written to/from memory
• I/O AR = input output address
• I/O BR = input output buffer register, the data

Memory is an array of bytes, each with its own address.

I/O devices constantly content for memory cycles, so every controller has a local buffer.

Interrupts #

Flow control with interrupts #

Here, the IO operation is concurrent with CPU processing.

Interrupts allow overlap of CPU and IO operations.

The interrupt vector is just a lookup table.

We allow the overlap of CPU and IO operations.