Vorlesung 12 Flashcards

1
Q

Intel’s single core assessment

A

power and area increase more rapidly than performance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Why multi-core processors?

A

POWER – Formeln auf Folie 60

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the evolution of processor performance

A

Currently, end of the line?, because only 3% processor performance increase can be achieved within a year.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

ARM LITTLE

A

most energy efficient application processor from ARM

  • simple architecture
  • in-order
  • 8 stage pipeline
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

ARM big

A

highest performance in mobile power envelope

  • complex architecture
  • out-of-order
  • multi-issue pipeline
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

When does one use ARM’s little?

A

“Always on, always connected” tasks like OS, UI activity

Maximum efficiency

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

When does one use ARM’s big

A

for best performance…

“demanding tasks” like browsers, gaming, content creation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is SGEMM/W?

A

single-precision general matrix multiplication per Watt

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the lesson learned from nature?

A

High power efficiency of human brain comes from specialization. More advanced societies have higher degree of specilization.

—–> dedicated architectures have a higher energy efficiency compared to general purpose chip types

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

what is the main task of the zFAS (zentrales Fahrassistenzsteuergerät)

A

Sensor fusion (signals from stereo cameras, radar, multi-axis acceleration sensors)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

what does a Field Programmable Gate Array (FPGA) consist of?

A

A set of programmable macro cells

A programmable interconnection network

Programmable input/outputs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

what are LUTs and what are they used for?

A

Look-up tables, used as function generators in memory FPGA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

characteristics of software

A

Large flexibility, easy to modify

„Easy“ to learn, many teaching courses

Minimal infrastructure necessary

Large amount of free and open source software

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

characteristics of hardware

A

Any software requires underlying hardware

Some functions can‘t be implemented in software, e.g., amplifier

Hardware is inherently parallel

Hardware engineering is based on physical laws

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

where does the problem of concurrency in software come from and how can it be solved?

A

programming languages are usually based on sequential models…

can be solved by having parallelized code

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are characteristics of the High-Performance-Computing (HPC) world?

A

almost unlimited resources

reduce computing time for large computation

but energy is now a big issue!

17
Q

Characteristics of the embedded world

A

restricted resources (memory, power,…)

heterogeneous architecture

real time and safety critical applications

18
Q

What are the limitations of concurrency in software

A

Not possible for every algorithm: no independent instructions

More resources do not necessarily reduce the computation time

Splitting the work in two does not mean dividing the time by 2

Using n processors for n tasks might waste resources without decreasing the
computation time

Speed
up depends on the algorithm and is never linear!
Rule of thumb: N processors
√𝑁performance increase

19
Q

What is the rule of thumb for the performance increase based on the number of processors?

A

N processors lead to a performance increase of √𝑁

20
Q

Amdahl’s Law formula and meaning

A

PPT

Amdahl’s Law describes performance gain by parallelization

21
Q

Difference between Amdahl’s and Gustafson’s Laws

A

Amdahl considered a fixed size problem, Gustafson a problem size that scales

22
Q

What are basic types of parallelization and give a quick example

A

Task parallelism: several independent tasks

data parallelism: split a work into small identical tasks

loop unrolling/pipelining

23
Q

What is the main challenge in parallelization?

A

Ensure determinism by adding lock mechanisms.

24
Q

What is the problem of parallelization techniques regarding the embedded world

A

many parallelization techniques not suitable for the embedded world:

Not suitable for safety critical applications

Real time requirements not taken into account

25
Q

What are the three domains of hardware design methodology

A

behavioral, structural, geometry

26
Q

What is a synthesis

A

mapping behaviour onto a structure

27
Q

Analysis

A

extracting behaviour from structure

28
Q

Generation

A

mapping structure onto geometry

29
Q

Extraction

A

extracting structure from geometry

30
Q

Synthesis techniques

A

Circuit synthesis: mapping transfer functions onto transistor structures

Logic synthesis: mapping Boolean equations onto gates

High Level synthesis: mapping RTL specifications onto automatons and datapaths

System synthesis: mapping a system specification onto a mixed HW/SW structure

31
Q

DSP (digital signal processing) Algorithms, steps of the transformation

A

Allocation:
selecting the number and types of resources
E.g. number of processors, memories, communication

Scheduling:
determines execution order of operations/processes

Mapping:
assigning operations/processes to resources

32
Q

HW/SW codesign flow

A

PPT

33
Q

HW/SW codesign

A

PPT

34
Q

ab 105 nicht drin, aber sieht hart useless aus

A

USELESS