Fault Tolerance Flashcards

1
Q

What are faults?

A

Unavoidable events

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

What is fault tolerance?

A

The aim to design a system in such a way that faults do not result in system failure

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

What is redundancy?

A

The use of additional elements within a system which would be required if the system was free of faults.

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

What is the triple modular redundancy (TMR) system?

A

A fault-tolerant system architecture that triplicate the processing elements and associated hardware , running them concurrently and uses voting mechanisms to achieve reliability by tolerating and correcting single faults.

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

What does a TMR system consist of?

A

3 identical hardware modules and one voting module.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is the output of a TMR’s system voting module?

A

Output of the majority of the modules.

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

What is temporal redundancy?

A

Used in order to tolerate or detect transient faults

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

What is the disadvantage of TMR?

A
  • Doesn’t cover design faults in the modules
  • If one module fails it is likely that identical modules fail at the same time
  • Helps only against random faults not systematic faults
  • Doesn’t help against simultaneous failures of multiple modules.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What does many fault-tolerance techniques rely on?

A

Detection of faults

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

What are the three methods for obtaining hardware fault tolerance?

A
  • Static redundancy
  • Dynamic redundancy
  • Hybrid approaches
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is static redundancy?

A

Static redundancy uses fault masking to hide faults, so that the system continues to work correctly even if a fault occurs.

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

What is dynamic redundancy?

A

Dynamic redundancy uses fault detection, to detect if a fault occurs and reconfigures the system in order to nullify the effects of faults.

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

What is hybrid approaches for hardware fault tolerance?

A

Using fault masking to prevent errors from propagating through the system and fault detect to reconfigure the system so that faulty units are removed

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

What may static redundancy use?

A

Fault masking and voting mechanisms.

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

What are single-point failures?

A

Vulnerabilities in a system where the malfunction/failure of a component can lead to a complete or partial breakdown of a system.

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

How can we avoid single point failures in the voting elements?

A

Triplicate the voting and pass the 3 output of the voting elements on to the next modules.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is N-Modular Redundancy (NMR)?

A

Uses N modules instead of 3 modules and voting among these

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What is the disadvantage of N-Modular Redundancy (NMR)?

A
  • Additional cost
  • Size
  • Weight
  • Power consumption
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What is the design of dynamic redundancy systems?

A

Use one unit as well as one or more standby systems.

20
Q

Which has more units Static or Dynamic redundancy?

A

Static redundancy has more units as all units have to be at least tripled.

21
Q

What is standby spare arrangement?

A

A standby spare arrangement has one module which is operated with some fault detection mechanism.

22
Q

What is a cold standy?

A

When the standby module is default off

23
Q

What is the disadvantage of a standby module in the standby spare agreement?

A
  • In case of a fault disruption is longer
  • Fault detection mechanism cannot make use of the data processed by the standby module
24
Q

What is a hot standy?

A

When the standby module processes the input but the output is ignored

25
Q

What is the disadvantage of hot standy?

A
  • More power consumption
  • Standby unit is subject to same operating stress as main module, if the main module can’t handle so can’t the hot standby module.
26
Q

What do self-checking pairs consists of?

A
  • one main module
  • one checking module
  • a comparator
27
Q

What does a comparator do in self-checking pairs?

A

Checks whether the output of the main modules coincides with the output of the checking module

28
Q

What is passed on from self-checking pairs?

A

The output of the main modules and the result of the comparison are passed onwards.

29
Q

Does self-checking pairs provide fault tolerance?

A

No, it provides error detection. But the output can be used in dynamic fault-tolerant systems

30
Q

What is hybrid redundancy?

A

Use of a combination of voting, fault detection and module switching.

31
Q

What is N-version programming?

A

N different versions of the software are written for the same spec and input data.

32
Q

What is the problem of N-version programming?

A
  • Development cost
  • Processing power
33
Q

What are N-version programming primarily used for?

A

Very critical application such as Airbus or space shuttle.

34
Q

What is Recovery block technique?

A

Based on acceptance tests, which are software versions of fault detection.

35
Q

What do acceptance test check?

A

The consistency of the output of one software module

36
Q

What are some examples of acceptance tests?

A
  • Check whether output is within boundaries
  • Check for run time errors
  • Check for excessive execution time
37
Q

What is the process of recovery blocks?

A
  • Execute the main module
  • Carry out acceptance test
  • If this fails switch to alternative module
  • Carry out acceptance test
  • if this fails switch again
  • etc
  • If everything fails raise an error
38
Q

What is the most cost effective way to achieve fault tolerance?

A

Use Fault-tolerant architectures.

39
Q

What are Fault-tolerant architectures?

A

For critical parts of the system we use non-computer-based mechanisms as
additional safe guards.

40
Q

What is the general problem for static, dynamic redundancy and hybrid approaches?

A

We can’t achieve redundancy when the components make the final
decision

41
Q

What is the problem with Static redudancy?

A
  • Have to maintain possible redundant components.
  • Lack of adaptability
42
Q

What is the problem with Dynamic redudancy?

A
  • Reconfigurations can cause delays and other issues
  • Reliable fault detection is needed
43
Q

What is the problem with hybrid approaches?

A
  • Complex to design
  • Have to be adaptable.
44
Q

What are the methods for Hardware Fault Tolerance?

A
  • Static Redundancy
  • Dynamic Redundancy
  • Hybrid Approaches
45
Q

What are the methods for Software Fault Tolerance?

A
  • N-version Programming
  • Recovery Blocks