Chapter 2

Question 1

Define

Pipelining

Answer 1

Processor moves data or instructions into a conceptual pipe with all stages of the pipe processing simultaneously

Question 2

Define

Branch Prediction

Branch Prediction

Answer 2

Processor looks ahead in the instruction code fetched from memory and predicts which branches, or groups of instructions, are likely to be processed next.

Question 3

Define

Superscalar execution

Answer 3

This is the ability to issue more than one instruction in every processor clock cycle. (In effect, multiple parallel
pipelines are used.)

Question 4

Define

Data flow analysis

Answer 4

Processor analyzes which instructions are dependent on each other’s results, or data, to create an optimized schedule of instructions

Question 5

Define

Speculative execution

Answer 5

Using branch prediction and data flow analysis, some processors speculatively execute instructions ahead of their actual appearance in the program execution, holding the results in temporary locations, keeping
execution engines as busy as possible.

Question 6

4 possible examples listed

Performance balance examples

Answer 6

Increase the number of bits that are retrieved at one time by making DRAMs “wider” rather than “deeper” and by using wide bus data paths.

Change the DRAM interface to make it more efficient by including a cache.

Reduce the frequency of memory access by incorporating increasingly complex and efficient cache structures between the processor and main memory.

Increase the interconnect bandwidth between processors and memory by using higher speed buses and a hierarchy of buses to buffer and structure data flow.

Question 7

How to increase hardware speed of processor

Solution with Results

Answer 7

Shrink logic gate size. More gates, packed more tightly, increase clock rate. Propagation time for signals reduced

Question 8

How to increase size and speed of cache

Solution

Answer 8

Dedicate part of processor chip

Question 9

List

Problems with clock speed and logic density

Answer 9

Power
RC delay
Memory latency and throughput

Question 10

Define

Advantages of multicore

Answer 10

Increase performance without increasing clock rate. Larger cache justified.

Question 11

Define

Many Integrated Core (MIC)

Answer 11

Involves a homogeneous collection of general purpose processors on a single chip.

Question 12

Define

Graphics Processing Unit (GPU)

Answer 12

Core designed to perform parallel operations on graphics data. Traditionally found on plug-in graphics card, used as vector processors

Question 13

Define

Amdahl’s law

Answer 13

Essentially, parallel performance allowed by multicores will only cause a great increase in performance if software is adapted to run parallel and exploit this power.

Question 14

Define

Little’s law

Answer 14

Average number of items in a queueing system equals the average rate at which items arrive multiplied by the time that an item spends in the system.

Question 15

List

The three common formulas used for calculating a mean

Answer 15

Arithmetic
Geometric
Harmonic

Question 16

Define

Arithmetic Mean (AM)

Answer 16

Appropriate if the sum of all the measurements is a meaningful and interesting value. Good for comparing the execution time performance of several systems.

Question 17

Define

Harmonic Mean (HM)

Answer 17

For situations where the execution rate is the more useful value of the system. This value is inversely proportional to the execution rate.

Question 18

Define

Geometric Mean (GM)

Answer 18

Gives equal weight to all values in a data set. Provides consistent results when measuring the relative performance of machines. Results are normalized to a reference machine.

Question 19

4 desirable characteristics of a benchmark program.

Answer 19

1. It is written in a high-level language
2. It is representative of a particular kind of programming domain or paradigm, such as systems programming, numerical programming, or commercial programming
3. It can be measured easily
4. It has wide distribution

Question 20

Define

System Performance Evaluation Corporation (SPEC)

Answer 20

Defines and maintains the best known collection of benchmark suites aimed at evaluating computer systems. Performance measurements are widely used for comparison and research purposes.

Question 21

Define

SPEC CPU2017

Answer 21

Best known SPEC benchmark suite, for applications doing more computation than I/O. Consists of 20 integer benchmarks and 23 float benchmarks. Contains over 11 million lines of code.

Question 22

Define

Peak metric

Answer 22

Enables users to attempt to optimize system performance by optimizing the compiler output.

Question 23

Define

Speed metric

Answer 23

A measurement of the time it takes to execute a compiled benchmark. Used for comparing the ability of a computer to complete single tasks.

Question 24

Define

Rate metric

Answer 24

Measurement of how many tasks a computer can accomplish in a certain time. AKA: throughput, capacity, rate measure.