Computer Organization and Architecture

Designing for Performance

11th Edition

Chapter 2
Performance Issues

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Designing for Performance
• The cost of computer systems continues to drop dramatically, while the performance and capacity
of those systems continue to rise equally dramatically
• Today’s laptops have the computing power of an IBM mainframe from 10 or 15 years ago
• Processors are so inexpensive that we now have microprocessors we throw away
• Desktop applications that require the great power of today’s microprocessor-based systems
include:
– Image processing
– Three-dimensional rendering
– Speech recognition
– Videoconferencing
– Multimedia authoring
– Voice and video annotation of files
– Simulation modeling
• Businesses are relying on increasingly powerful servers to handle transaction and database
processing and to support massive client/server networks that have replaced the huge mainframe
computer centers of yesteryear
• Cloud service providers use massive high-performance banks of servers to satisfy high-volume,
high-transaction-rate applications for a broad spectrum of clients

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Microprocessor Speed
Techniques built into contemporary processors include:
Pipelining • Processor moves data or instructions into a conceptual pipe
with all stages of the pipe processing simultaneously

• Processor looks ahead in the instruction code fetched from

Branch prediction memory and predicts which branches, or groups of
instructions, are likely to be processed next

• This is the ability to issue more than one instruction in every

Superscalar execution processor clock cycle. (In effect, multiple parallel pipelines
are used.)

• Processor analyzes which instructions are dependent on each

Data flow analysis other’s results, or data, to create an optimized schedule of
instructions

• Using branch prediction and data flow analysis, some

processors speculatively execute instructions ahead of their
Speculative execution actual appearance in the program execution, holding the
results in temporary locations, keeping execution engines as
busy as possible

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Performance Balance
• Adjust the organization and Increase the number of
architecture to compensate bits that are retrieved at
one time by making
DRAMs “wider” rather
for the mismatch among the than “deeper” and by
using wide bus data
capabilities of the various paths

components
Reduce the frequency of
• Architectural examples memory access by
incorporating
include: increasingly complex and
efficient cache structures
between the processor
and main memory

Change the DRAM Increase the interconnect

interface to make it more bandwidth between
efficient by including a processors and memory
cache or other buffering by using higher speed
scheme on the DRAM buses and a hierarchy of
chip buses to buffer and
structure data flow

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Figure 2.1

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Improvements in Chip Organization and
Architecture
• Increase hardware speed of processor
– Fundamentally due to shrinking logic gate size
▪ More gates, packed more tightly, increasing clock rate
▪ Propagation time for signals reduced

• Increase size and speed of caches

– Dedicating part of processor chip
▪ Cache access times drop significantly

• Change processor organization and architecture

– Increase effective speed of instruction execution
– Parallelism

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Problems with Clock Speed and Logic
Density

• Power
– Power density increases with density of logic and clock speed
– Dissipating heat

• RC delay
– Speed at which electrons flow limited by resistance and capacitance of
metal wires connecting them
– Delay increases as the RC product increases
– As components on the chip decrease in size, the wire interconnects
become thinner, increasing resistance
– Also, the wires are closer together, increasing capacitance

• Memory latency and throughput

– Memory access speed (latency) and transfer speed (throughput) lag
processor speeds
Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Figure 2.2

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
 The use of multiple processors on

Multicore
the same chip provides the
potential to increase performance
without increasing the clock rate

Strategy is to use two simpler

processors on the chip rather than
one more complex processor

With two processors larger caches

are justified

As caches became larger it made

performance sense to create two
and then three levels of cache on a
chip

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Many Integrated Core (MIC)
Graphics Processing Unit (GPU)
MIC • GP
U
• Leap in performance as well • Core designed to perform
as the challenges in parallel operations on graphics
developing software to data
exploit such a large number
of cores • Traditionally found on a plug-in
graphics card, it is used to
• The multicore and MIC encode and render 2D and 3D
strategy involves a graphics as well as process
homogeneous collection of video
general purpose processors
on a single chip • Used as vector processors for
a variety of applications that
require repetitive computations

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Amdahl’s Law
• Gene Amdahl
• Deals with the potential speedup of a program using
multiple processors compared to a single processor
• Illustrates the problems facing industry in the
development of multi-core machines
– Software must be adapted to a highly parallel execution
environment to exploit the power of parallel processing

• Can be generalized to evaluate and design technical

improvement in a computer system

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Figure 2.3

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Figure 2.4

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Little’s Law
• Fundamental and simple relation with broad applications
• Can be applied to almost any system that is statistically in steady
state, and in which there is no leakage
• Queuing system
– If server is idle an item is served immediately, otherwise an arriving item joins
a queue
– There can be a single queue for a single server or for multiple servers, or
multiple queues with one being for each of multiple servers

• Average number of items in a queuing system equals the average

rate at which items arrive multiplied by the time that an item
spends in the system
– Relationship requires very few assumptions
– Because of its simplicity and generality it is extremely useful

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Figure 2.5

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Table 2.1 Performance Factors and System Attributes

Ic p m k 

Instruction set architecture X X

Compiler technology X X X

Processor implementation X X

Cache and memory hierarchy X X

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Calculating the Mean

The three
The use of benchmarks to compare common
systems involves calculating the formulas used
mean value of a set of data points
related to execution time for calculating a
mean are:

• Arithmetic
• Geometric
• Harmonic
Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Figure 2.6

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Arithmetic Mean
 An Arithmetic Mean (AM) is an appropriate measure
if the sum of all the measurements is a meaningful
and interesting value

 The AM is a good candidate for comparing the execution

time performance of several systems
For example, suppose we were interested in using a system
for large-scale simulation studies and wanted to evaluate several alternative
products. On each system we could run the simulation multiple times with
different input values for each run, and then take the average execution time
across all runs. The use of
multiple runs with different inputs should ensure that the results are not
heavily biased by some unusual feature of a given input set. The AM of all
the runs is a good measure of the system’s performance on simulations, and a
good number to use for system comparison.

 The AM used for a time-based variable, such as program execution time, has
the important property that it is directly proportional to the total time
 If the total time doubles, the mean
Copyright value
© 2019, 2016,doubles
2013 Pearson Education, Inc. All Rights Reserved
Table 2.2
A Comparison of Arithmetic and Harmonic Means for Rates
Computer Computer Computer Computer Computer Computer
A time B time C time A rate B rate C rate
(secs) (secs) (secs) (MFLOPS) (MFLOPS) (MFLOPS)

Program 1
2.0 1.0 0.75 50 100 133.33
(108 FP ops)

Program 1
0.75 2.0 4.0 133.33 50 25
(108 FP ops)

Total
execution 2.75 3.0 4.75 – – –
time

Arithmetic
mean of 1.38 1.5 2.38 – – –
times

Inverse
of total
0.36 0.33 0.21 – – –
execution
time (1/sec)

Arithmetic
mean of – – – 91.67 75.00 79.17
rates

Harmonic
mean of – – – 72.72 66.67 42.11
rates

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Table 2.3
A Comparison of Arithmetic and Geometric Means for Normalized
Results
(a) Results normalized to Computer A

Computer A time Computer B time Computer C time

Program 1 2.0 (1.0) 1.0 (0.5) 0.75 (0.38)

Program 2 0.75 (1.0) 2.0 (2.67) 4.0 (5.33)

Total execution time 2.75 3.0 4.75

Arithmetic mean of
1.00 1.58 2.85
normalized times
Geometric mean of
1.00 1.15 1.41
normalized times

(a) Results normalized to Computer B

Computer A time Computer B time Computer C time

Program 1 2.0 (2.0) 1.0 (1.0) 0.75 (0.75)

Program 2 0.75 (0.38) 2.0 (1.0) 4.0 (2.0)

Total execution time 2.75 3.0 4.75

Arithmetic mean of
1.19 1.00 1.38
normalized times
Geometric mean of
0.87 1.00 1.22
normalized times

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Table 2.4
Another Comparison of Arithmetic and Geometric Means for
Normalized Results
(a) Results normalized to Computer A

Computer A time Computer B time Computer C time

Program 1 2.0 (1.0) 1.0 (0.5) 0.20 (0.1)

Program 2 0.4 (1.0) 2.0 (5.0) 4.0 (10.0)

Total execution time 2.4 3.00 4.2

Arithmetic mean of
1.00 2.75 5.05
normalized times
Geometric mean of
1.00 1.58 1.00
normalized times

(a) Results normalized to Computer B

Computer A time Computer B time Computer C time

Program 1 2.0 (2.0) 1.0 (1.0) 0.20 (0.2)

Program 2 0.4 (0.2) 2.0 (1.0) 4.0 (2.0)

Total execution time 2.4 3.0 4.2

Arithmetic mean of
1.10 1.00 1.10
normalized times
Geometric mean of
0.63 1.00 0.63
normalized times

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Benchmark Principles
• Desirable characteristics of a benchmark
program:

1. It is written in a high-level language, making it portable

across different machines
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

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
System Performance Evaluation
Corporation (SPEC)
• Benchmark suite
– A collection of programs, defined in a high-level language
– Together attempt to provide a representative test of a computer in a
particular application or system programming area

– SPEC
– An industry consortium
– 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

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
SPEC CPU2017
• Best known SPEC benchmark suite
• Industry standard suite for processor intensive applications
• Appropriate for measuring performance for applications that
spend most of their time doing computation rather than I/O
• Consists of 20 integer benchmarks and 23 floating-point
benchmarks written in C, C++, and Fortran
• For all of the integer benchmarks and most of the floating-
point benchmarks, there are both rate and speed benchmark
programs
• The suite contains over 11 million lines of code

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
 Rate Speed Language Kloc Application Area

500.perlbench_r 600.perlbench_s C 363 Perl interpreter

502.gcc_r 602.gcc_s C 1304 GNU C compiler

505.mcf_r 605.mcf_s C 3 Route planning

520.omnetpp_r 620.omnetpp_s C++ 134 Discrete event simulation - computer

network
Table 2.5
523.xalancbmk_r 623.xalancbmk_s C++ 520 XML to HTML conversion via XSLT (A)

525.x264_r 625.x264_s C 96 Video compression SPEC

531.deepsjeng_r 631.deepsjeng_s C++ 10 AI: alpha-beta tree search (chess)
CPU2017
Benchmarks
541.leela_r 641.leela_s C++ 21 AI: Monte Carlo tree search (Go)

548.exchange2_r 648.exchange2_s Fortran 1 AI: recursive solution generator (Sudoku)

557.xz_r 657.xz_s C 33 General data compression

Kloc = line count (including comments/whitespace) for source files used in a build/1000 (Table can be found on page 61 in the textbook.)
Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
 Rate Speed Language Kloc Application Area

503.bwaves_r 603.bwaves_s Fortran 1 Explosion modeling

507.cactuBSSN_r 607.cactuBSSN_s C++, C, 257 Physics; relativity
Fortran

508.namd_r C++, C 8 Molecular dynamics

510.parest_r C++ 427 Biomedical imaging; optical tomography
with finite elements
Table 2.5
(B)
511.povray_r C++ 170 Ray tracing
519.ibm_r 619.ibm_s C 1 Fluid dynamics SPEC
521.wrf_r 621.wrf_s Fortran, C 991 Weather forecasting
CPU2017
526.blender_r C++ 1577 3D rendering and animation
Benchmarks
527.cam4_r 627.cam4_s Fortran, C 407 Atmosphere modeling
628.pop2_s Fortran, C 338 Wide-scale ocean modeling (climate
level)

538.imagick_r 638.imagick_s C 259 Image manipulation

544.nab_r 644.nab_s C 24 Molecular dynamics
549.fotonik3d_r 649.fotonik3d_s Fortran 14 Computational electromagnetics

554.roms_r 654.roms_s Fortran 210 Regional ocean modeling.

Kloc = line count (including comments/whitespace) for source files used in a build/1000 (Table can be found on page 61 in the textbook.)
Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
 Base Peak

Seconds Rate Seconds Rate

Benchmark
1141 1070 933 1310
500.perlbench_r Table 2.6
1303 835 1276 852
502.gcc_r
1433 866 1378 901
SPEC
505.mcf_r
1664 606 1634 617
CPU 2017
520.omnetpp_r Integer
722 1120 713 1140
Benchmarks
523.xalancbmk_r for HP
525.x264_r
655 2053 661 2030 Integrity
604 1460 597 1470 Superdome X
531.deepsjeng_r

541.leela_r
892 1410 896 1420 (a) Rate Result
833 2420 770 2610 (768 copies)
548.exchange2_r

870 953 863 961

557.xz_r
© 2018 Pearson Education, Inc., Hoboken, NJ. All rights reserved.
(Table can be found on page 64 in the textbook.)
Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
 Base Peak

Seconds Ratio Seconds Ratio

Benchmark
358 4.96 295 6.01 Table 2.6
600.perlbench_s

602.gcc_s
546 7.29 535 7.45 SPEC
605.mcf_s
866 5.45 700 6.75 CPU 2017
276 5.90 247 6.61 Integer
620.omnetpp_s
Benchmarks
188 7.52 179 7.91 for HP
623.xalancbmk_s
Integrity
625.x264_s
283 6.23 271 6.51 Superdome X
407 3.52 343 4.18
631.deepsjeng_s
(b) Speed
641.leela_s
469 3.63 439 3.88
Result
329 8.93 299 9.82
(384 threads)
648.exchange2_s

2164 2.86 2119 2.92

657.xz_s

(Table can be found on page 64 in the textbook.)

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Terms Used in SPEC Documentation
• Benchmark • Peak metric
– A program written in a high-level – This enables users to attempt to
language that can be compiled and optimize system performance by
executed on any computer that optimizing the compiler output
implements the compiler • Speed metric
• System under test – This is simply a measurement of the
– This is the system to be evaluated time it takes to execute a compiled
• Reference machine benchmark
– This is a system used by SPEC to • Used for comparing the ability of a
establish a baseline performance for all computer to complete single tasks
benchmarks • Rate metric
▪ Each benchmark is run and – This is a measurement of how many
measured on this machine to tasks a computer can accomplish in a
establish a reference time for that certain amount of time
benchmark • This is called a throughput, capacity,
• Base metric or rate measure
– These are required for all reported • Allows the system under test to
results and have strict guidelines for execute simultaneous tasks to take
compilation advantage of multiple processors
Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Figure 2.7

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
 Benchmark Seconds Energy (kJ) Average Power Maximum Power
(W) (W)

1774 1920 1080 1090

600.perlbench_s

3981 4330 1090 1110

602.gcc_s

605.mcf_s
4721 5150 1090 1120
Table 2.7
1630 1770 1090 1090
620.omnetpp_s SPECspeed
2017_int_base
1417 1540 1090 1090
Benchmark
623.xalancbmk_s Results for
Reference
1764 1920 1090 1100
625.x264_s Machine (1
1432 1560 1090 1130 thread)
631.deepsjeng_s

1706 1850 1090 1090

641.leela_s
2939 3200 1080 1090
648.exchange2_s

6182 6730 1090 1140

657.xz_s
(Table can be found on page 66 in the textbook.)
Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Summary • Performance
• Concepts
Chapter 2
• Designing for performance • Basic measures of computer
– Microprocessor speed performance
– Performance balance – Clock speed
– Improvements in chip – Instruction execution rate
organization and
• Calculating the mean
architecture
– Arithmetic mean
• Multicore
– Harmonic mean
• MICs
– Geometric mean
• GPGPUs
• Amdahl’s Law • Benchmark principles
• Little’s Law • SPEC benchmarks

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved
Copyright

This work is protected by United States copyright laws and is provided solely
for the use of instructions in teaching their courses and assessing student
learning. dissemination or sale of any part of this work (including on the
World Wide Web) will destroy the integrity of the work and is not permit-
ted. The work and materials from it should never be made available to
students except by instructors using the accompanying text in their
classes. All recipients of this work are expected to abide by these
restrictions and to honor the intended pedagogical purposes and the needs of
other instructors who rely on these materials.

Copyright © 2019, 2016, 2013 Pearson Education, Inc. All Rights Reserved