1.1.2 Types of Processors

0.0(0)
Studied by 2 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/17

flashcard set

Earn XP

Description and Tags

Last updated 3:53 PM on 11/21/22
Name
Mastery
Learn
Test
Matching
Spaced
Call with Kai

No analytics yet

Send a link to your students to track their progress

18 Terms

1
New cards
CISC
Complex Instruction Set Computer

Tries to reduce the number of instructions that a program has by combining many simple instructions into a single complex one.
2
New cards
Attributes of CISC
- Large instruction set
- Complex instructions
- Clock ticks slightly slower
- Each instruction make take many cycles and so program runs slower
- Less use of RAM
- Variable number of bytes and so instructions are larger than size of word
- Hardware is more complex and bigger
- More transistors so consumes more power and requires heat sink/fan
- More transistors = more cost
- Used in desktops, PCs and Laptops
3
New cards
RISC
Reduced Instruction Set Computer

They have more instuctions, but they reduce the number of cycles that an instruction takes to perform

- Programs run faster due to smaller instructions
- Small instruction set
4
New cards
Attributes of RISC
- Limited number of instructions
- Simple processor design
- Each instruction takes one cycle so program runs quicker
- Heavy use of RAM
- Fixed number of bytes
- Hardware is simpler and smaller in design
- Used fewer transistors, so doesn't get as hot
- less transistors = less cost
- Used in smartphones and tablets
5
New cards
Concurrent
At the same time
6
New cards
Pipeline
SISD
Inside CPU
Overlaps the Fetch-Decode-Execute cycle
Can't be done with branching statements and things that require human input
7
New cards
SISD
Single Instruction Single Data
1 instruction on one piece of data
8
New cards
Array processor
SIMD
Graphics cards, games consoles
A single instruction is issued by the control unit and is applied to a number of data sets at the same time
9
New cards
SIMD
Single Instruction Multiple Data
Same instruction to lots of pieces of data
10
New cards
Multi-core
MIMD
Super computers, modern multi-core chips
11
New cards
MIMD
Multiple Instruction Multiple Data
Lots of different instructions to lots of different pieces of data
12
New cards
Advantages of parallel processing
-Faster when handling large amounts of data, with each data set requiring the same processing (array and multicore methods)
-Not limited by the bus transfer rate (no von-Neumann bottleneck)
-Can make maximum use of CPU (pipeline method) in spite of the bottleneck

The GPU does not speed up the CPU, but it takes some of the calculations away from the CPU so that it can do other calculations, speeding up the system
13
New cards
Disadvantages of Parallel Processing
-Only certain types of data are suitable for parallel processing. Data that relies on the result of a previous operation cannot be made parallel, for parallel processing, each data set must be independent of each other.
•More costly in terms of hardware - multiple processing blocks needed, this applies to all three methods
14
New cards
Co-Processor
Especially designed to carry out floating point calculations extremely quickly and works with the CPU on the motherboard.
Advantage:
Calculation (and hence performance) is much faster.
Disadvantages:
- more expensive,
- requires more motherboard space
- takes more power.
- only has a limited range of functionS
15
New cards
GPU
Graphics Processing Unit which is designed to handle graphics and video much faster than the CPU itself
They have instruction sets which relate specifically to graphics processing
Has 1000s of cores designed for handling multiple tasks at the same time
Used as a co-processor
16
New cards

Co-Processing VS Parallel Processing

  • Both involve more than one processor working together to perform a single job with each job being split into smaller tasks.

  • Parallel processors can use any processor for a task -whereas co-processors only perform specific types of task e.g. floating point calculations. The task is only completed more quickly in a co-processor if it is a suitable task.

  • Parallel processors need a more complex OS than co-processors as it needs to control how tasks are broken down and spread across all available cores

17
New cards
GPU disadvantages
• expensive.
•do not improve CPU performance on all tasks, only ones related to graphics and video.
•use a lot of power needs a more powerful power supply.
• need a cooling fan = noisy .
18
New cards
Three types of parallel processing
Pipelining, array processor, multicore