COMP1860 Lecture 7.2 Building our Digital World: Computer Systems & Architecture

0.0(0)
studied byStudied by 0 people
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
Card Sorting

1/16

flashcard set

Earn XP

Description and Tags

Flashcards covering key vocabulary and concepts from the lecture on I/O, including programmed I/O, interrupt-driven I/O, DMA, and DCA.

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

17 Terms

1
New cards

I/O Module

A hardware component that manages data exchange between external devices and the CPU, handling control, status, and data signals.

2
New cards

Peripheral

An external device that connects to a computer system via an I/O module, classified as human-readable, machine-readable, or communication.

3
New cards

Control Signals

Signals that determine device functions, such as READ, WRITE, or device-specific tasks.

4
New cards

Data Signals

Bits exchanged between the device and the I/O module.

5
New cards

Status Signals

Signals indicating the device's state, such as READY/NOT-READY.

6
New cards

Programmed I/O

An I/O technique where the processor controls I/O operations and waits for completion, continuously checking the status register.

7
New cards

Memory-Mapped I/O

An I/O scheme where I/O devices share a single address space with memory, allowing the same machine instructions for memory and I/O operations.

8
New cards

Isolated I/O

An I/O scheme with separate address spaces and buses for memory and I/O, accessed through specific I/O commands.

9
New cards

Interrupt-Driven I/O

An I/O technique where the processor issues an I/O command and continues execution; the I/O module interrupts when the data is ready.

10
New cards

Direct Memory Access (DMA)

An I/O technique where the I/O module transfers data directly to memory, bypassing the processor for improved efficiency.

11
New cards

Direct Cache Access (DCA)

An I/O technique that allows I/O devices to access the last-level cache (LLC) directly, reducing memory latency and improving data transfer efficiency.

12
New cards

Cycle Stealing

A technique used by DMA where it accesses the system bus without disrupting the processor, by suspending the processor briefly.

13
New cards

Ring Interconnect

An architecture that links cores, last-level cache, PCIe, and memory controller in a ring structure, enabling efficient scaling for multi-core systems.

14
New cards

DDIO (Direct Data I/O)

An implementation of DCA in Intel Xeon processors where incoming and outgoing packets are placed directly into the cache instead of memory.

15
New cards

I/O Channel

A specialized processor that executes I/O instructions independently of the CPU.

16
New cards

Selector Channel

An I/O channel that manages multiple high-speed devices, transferring data with one device at a time.

17
New cards

Multiplexor Channel

An I/O channel that handles multiple devices simultaneously, interleaving data streams from low-speed or high-speed devices.