L2: Architectures

Question 1

What is Flynns Taxonomy (1 : 5)

Answer 1

Classifies multi-processor computer architectures according to two independent dimensions.
Instruction Stream
Data Stream

Two possible states:
Single or Multiple.

4 Classifications.

Question 2

SISD - Single instruction, Single Data. What is it, features ? ( 4 : 6 )

Answer 2

Serial Computer
Single instruction
- Only one instruction stream is being acted on
Single data
- One data stream is used during one clock cycle

• Good for real time apps
• Traditional single processor/core CPU

Question 3

SIMD - Single instruction, multiple data. What is it, features? ( 4 : 7 )

Answer 3

Parallel computer
Single Instruction
- All processing units execute same instruction at a clock cycle
Multi data
- Each unit can operate on a different data element

• Modern CPU and GPU
• Best for problems with high degree of regularity
• OpenCL

Question 4

MIMD - What is it, features ? ( 4: 8)

Answer 4

A type of parallel computer
Multiple Instructions
• every processor may be executing a different instruction stream
Multiple Data
• every processor may be working with a different data stream
Notes
• most common type of parallel computer, multi-core CPUs, computing clusters and grids
• many MIMD architectures also include SIMD execution sub-components

Question 5

MISD - What is it, features? (4 : 9 )

Answer 5

A type of parallel computer
Multiple Instruction
• each processing unit operates on the data independently via separate instruction streams
Single Data
• a single data stream is fed into multiple processing units
Notes
• multiple cryptography algorithms attempting to crack a single coded message
• very uncommon architecture and rare applications

Question 6

Two different memory architectures for SIMD/MIMD (2 : 10)

Answer 6

Shared
• processors operate
independently but share
the same memory –
global address space
• changes in memory by
one processor are visible
to all other processors

Distributed
• processors operate
independently but have
their own local memory
• memory addresses in
one processor do not
map to another
processor – no global
address space

Question 7

Pros and Cons of Shared and Distributed Memory Architectures (8 : 11 )

Answer 7

Shared
• Pros
• global address space is easy
to use/program
• data sharing between tasks
is fast due to proximity of
memory to CPUs
• Cons
• adding more CPUs can
increase traffic on shared
memory-CPU path
• programmer responsible for
synchronization to ensure
"correct" access to global
memory

Distributed

• Pros
• each processor can rapidly
access its own memory
without interference
• memory is scalable; increase
number of processors and
size of memory increases
• Cons
• programmer responsible for
data communication
between processors
• non-uniform memory access
times

Question 8

What is a GPU ? ( 4 : 13)

Answer 8

-Designed for manipulating computer
graphics and image processing
-Highly parallel structure
-More efficient than CPUs when processing
large blocks of visual data in parallel
-Different realizations
• dedicated expansion video card
• integrated into the CPU die
• embedded on motherboard

Question 9

What is a GPGPU? ( 1 : 15)

Answer 9

General Purpose GPU.
Application of GPU for applications other than
graphics, with large datasets and complex
computations
• everywhere where large vectors/matrices are used
• physics simulation, AI, weather forecasting
Required hardware changes (shaders, texture units,
FP arithmetics) so that standard code could be
executed.
Also software frameworks hiding the
graphics-specific functionality from the
programmer (e.g. CUDA, OpenCL).