chapter 1 Flashcards

1
Q

what are the microprocessor techniques?

A
  1. pipelining
  2. branch prediction
  3. data flow analysis
  4. superscalar execution
  5. speculative execution
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

explain pipelining

A

W - execute a series of instruction in an orderly manner

Stages 1. fetch 2. decode 3. execute 4. writeback

types 1. arithmetic - problems divide into sub problems
2. instructions - read next instruction & execute previous instruction

pros - reduce processor’s cycle time
cons - stalling time

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

explain superscalar execution

A

W - execute multiple instructions during a clock cycle in various execution units

H 1. fetch unit = fetch & read instructions
2. dispatch unit = decodes & error checking
3. 1 floating point & 1 integer point, dispatch at the same clock cycle

pros - keeps both units busy
- increase performance

cons - 1 multiple errors (scheduling problem)

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

explain data flow analysis

A

W - arrange a series of instructions in a sequential manner

H - 1. initialize = data flow values
2. propagation = propagate via data flow graph by analyzing relationships between statements
3. analyzing = error checking & program optimization
4. iteration = iterate propagation to prevent value alteration
5. output = summarize analysis & program variables

pros - error detections & code optimization

cons - complex & false positive

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

explain speculative execution

A

W - utilize idle processing units to prefetch data & executes them ahead

H 1. data flow analysis
2. pipelining
3. out of order execution

pros - increase performance

cons - decrease performance if components are busy

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

explain branch prediction

A

W - predict branch conditions and store them for future uses

H - obtain historical data to make predictions
- 2-level adaptive = dynamic adjust prediction based on recent data

pros - reduce delay & increase performance

cons - complex & power consumption
2. incorrect prediction penalty

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

explain von Neuman’s principle

A
  • the basis of computer system
  • main memory = store data & instructions
  • cpu = perform arithmetic & logic operations
  • pc = keeps track & fetch next instruction
  • io = operations of io devices
How well did you know this?
1
Not at all
2
3
4
5
Perfectly