8/13/2019 ACA UNit 1

1/29

1

Advanced Computer

Architecture

UnitI

Fundamentals

Of Computer Design

8/13/2019 ACA UNit 1

2/29

2

Introduction

1.1 Introduction

1.2 The Task of a Computer Designer

1.3 Technology and Computer Usage Trends

1.4 Cost and Trends in Cost

1.5 Measuring and Reporting Performance

1.6 Quantitative Principles of Computer Design

8/13/2019 ACA UNit 1

3/29

3

Whats Computer Architecture?

The attributes of a [computing] system as seen by the

programmer, i.e., the conceptual structure and functional

behavior, as distinct from the organization of the data

flows and controls the logic design, and the physicalimplementation.

SOFTWARE

8/13/2019 ACA UNit 1

4/29

4

Whats Computer Architecture?

1950s to 1960s: Computer Architecture Course

Computer Arithmetic.

1970s to mid 1980s: Computer Architecture Course

Instruction Set Design, especially ISA appropriate forcompilers. (What well do in Chapter 2)

1990s to 2000s: Computer Architecture Course

Design of CPU, memory system, I/O system,

Multiprocessors. (All evolving at a tremendous rate!)

8/13/2019 ACA UNit 1

5/29

5

The Task of a

Computer DesignerEvaluate Existing

Systems for

Bottlenecks

Simulate NewDesigns and

Organizations

Implement Next

Generation System

Technology

Trends

Benchmarks

Workloads

Implementat ion

Complexi ty

8/13/2019 ACA UNit 1

6/29

6

Technology and

Computer Usage Trends

Similarly, Computer Architecture is about

working within constraints: What will the market buy?

Cost/Performance

Tradeoffs in materials and processes

When building a Cathedral numerous

very practical considerations need to

be taken into account:

available materials

worker skills willingness of the client to pay the

price.

8/13/2019 ACA UNit 1

7/29

7

TrendsGordon Moore (Founder of Intel) observed in 1965 that the number of

transistors that could be crammed on a chip doubles every year.

This has CONTINUED to be true since then.Transistors Per Chip

1.E+03

1.E+04

1.E+05

1.E+06

1.E+07

1.E+08

1970 1975 1980 1985 1990 1995 2000 2005

4004

Power PC 601486

386

80286

8086

Pentium

Pentium Pro

Pentium II

Power PC G3

Pentium 3

8/13/2019 ACA UNit 1

8/29

8

TrendsProcessor performance, as measured by the SPEC benchmark has

also risen dramatically.

0

1000

2000

3000

4000

5000

87

88

89

90

91

92

93

94

95

96

97

98

99

2000

DEC Alpha 21264/600

DEC Alpha 5/500

DEC Alpha 4/266

DEC

AXP/

500Sun

-4/260

IBM

RS/

6000

MIPS

M2000

Alpha 6/833

8/13/2019 ACA UNit 1

9/29

9

TrendsMemory Capacity (and Cost) have changed dramatically

in the last 20 years.

size

Year

1000

10000

100000

1000000

10000000

100000000

1000000000

1970 1975 1980 1985 1990 1995 2000

year size(Mb) cyc time

1980 0.0625 250 ns

1983 0.25 220 ns

1986 1 190 ns

1989 4 165 ns

1992 16 145 ns

1996 64 120 ns

2000 256 100 ns

8/13/2019 ACA UNit 1

10/29

10

Trends

Based on SPEED, the CPU has increased dramatically, but memory

and disk have increased only a little. This has led to dramaticchanged in architecture, Operating Systems, and Programming

practices.

Capacity Speed (latency)Logic 2x in 3 years 2x in 3 years

DRAM 4x in 3 years 2x in 10 years

Disk 4x in 3 years 2x in 10 years

8/13/2019 ACA UNit 1

11/29

11

Measuring And

Reporting Performance

This section talks about:

1. Metricshow do we describe in a numerical way theperformance of a computer?

2. What tools do we use to find those metrics?

8/13/2019 ACA UNit 1

12/29

12

Metrics

Time to run the task (ExTime) Execution time, response time, latency

Tasks per day, hour, week, sec, ns (Performance) Throughput, bandwidth

Plane

Boeing 747

BAD/SudConcodre

Speed

610 mph

1350 mph

DC to Paris

6.5 hours

3 hours

Passengers

470

132

Throughput

(pmph)

286,700

178,200

8/13/2019 ACA UNit 1

13/29

13

Metrics - Comparisons

"X is n times faster than Y" means

ExTime(Y) Performance(X)

--------- = ---------------

ExTime(X) Performance(Y)

Speed of Concorde vs. Boeing 747

Throughput of Boeing 747 vs. Concorde

8/13/2019 ACA UNit 1

14/29

14

Metrics - ComparisonsPat has developed a new product, "rabbit" about which she wishes to determine

performance. There is special interest in comparing the new product, rabbit to the

old product, turtle, since the product was rewritten for performance reasons. (Pat

had used Performance Engineering techniques and thus knew that rabbit was

"about twice as fast" as turtle.) The measurements showed:

Performance Comparisons

Product Transactions / second Seconds/ transaction Seconds to process transaction

Turtle 30 0.0333 3Rabbit 60 0.0166 1

Which of the following statements reflect the performance comparison of rabbit and

turtle?

o Rabbit is 100% faster than turtle.o Rabbit is twice as fast as turtle.

o Rabbit takes 1/2 as long as turtle.

o Rabbit takes 1/3 as long as turtle.

o Rabbit takes 100% less time than turtle.

o Rabbit takes 200% less time than turtle.o Turtle is 50% as fast as rabbit.

o Turtle is 50% slower than rabbit.

o Turtle takes 200% longer than rabbit.

o Turtle takes 300% longer than rabbit.

8/13/2019 ACA UNit 1

15/29

15

Metrics - Throughput

Compiler

Programming

Language

Application

DatapathControl

Transistors Wires Pins

ISA

Function Units

(millions) of Instructions per second: MIPS

(millions) of (FP) operations per second: MFLOP/s

Cycles per second (clock rate)

Megabytes per second

Answers per month

Operations per second

8/13/2019 ACA UNit 1

16/29

16

Methods For Predicting

Performance Benchmarks, Traces, Mixes

Hardware: Cost, delay, area, power estimation

Simulation (many levels)

ISA, RT, Gate, Circuit

Queuing Theory

Rules of Thumb

Fundamental Laws/Principles

8/13/2019 ACA UNit 1

17/29

17

Benchmarks

First Round 1989

10 programs yielding a single number (SPECmarks)

Second Round 1992

SPECInt92 (6 integer programs) and SPECfp92 (14 floating point programs)

Compiler Flags unlimited. March 93 of DEC 4000 Model 610:

spice: unix.c:/def=(sysv,has_bcopy,bcopy(a,b,c)=memcpy(b,a,c)

wave5: /ali=(all,dcom=nat)/ag=a/ur=4/ur=200

nasa7: /norecu/ag=a/ur=4/ur2=200/lc=blas

Third Round 1995 new set of programs: SPECint95 (8 integer programs) and SPECfp95 (10 floating

point)

benchmarks useful for 3 years

Single flag setting for all programs: SPECint_base95, SPECfp_base95

SPEC: System Performance Evaluation

Cooperative

8/13/2019 ACA UNit 1

18/29

18

BenchmarksCINT2000 (Integer Component of SPEC CPU2000):

Program Language What Is It

164.gzip C Compression

175.vpr C FPGA Circuit Placement and Routing

176.gcc C C Programming Language Compiler

181.mcf C Combinatorial Optimization

186.crafty C Game Playing: Chess

197.parser C Word Processing

252.eon C++ Computer Visualization

253.perlbmk C PERL Programming Language

254.gap C Group Theory, Interpreter255.vortex C Object-oriented Database

256.bzip2 C Compression

300.twolf C Place and Route Simulator

http://www.spec.org/osg/cpu2000/CINT2000/

8/13/2019 ACA UNit 1

19/29

19

BenchmarksCFP2000 (Floating Point Component of SPEC

CPU2000):Program Language What Is It168.wupwise Fortran 77 Physics / Quantum Chromodynamics

171.swim Fortran 77 Shallow Water Modeling

172.mgrid Fortran 77 Multi-grid Solver: 3D Potential Field

173.applu Fortran 77 Parabolic / Elliptic Differential Equations

177.mesa C 3-D Graphics Library178.galgel Fortran 90 Computational Fluid Dynamics

179.art C Image Recognition / Neural Networks

183.equake C Seismic Wave Propagation Simulation

187.facerec Fortran 90 Image Processing: Face Recognition

188.ammp C Computational Chemistry

189.lucas Fortran 90 Number Theory / Primality Testing191.fma3d Fortran 90 Finite-element Crash Simulation

200.sixtrack Fortran 77 High Energy Physics Accelerator Design

301.apsi Fortran 77 Meteorology: Pollutant Distribution

8/13/2019 ACA UNit 1

20/29

20

Benchmarks Sample Results ForSpecINT2000

Base Base Base Peak Peak PeakBenchmarks Ref Time Run Time Ratio Ref Time Run Time Ratio

164.gzip 1400 277 505* 1400 270 518*

175.vpr 1400 419 334* 1400 417 336*

176.gcc 1100 275 399* 1100 272 405*

181.mcf 1800 621 290* 1800 619 291*

186.crafty 1000 191 522* 1000 191 523*

197.parser 1800 500 360* 1800 499 361*

252.eon 1300 267 486* 1300 267 486*

253.perlbmk 1800 302 596* 1800 302 596*

254.gap 1100 249 442* 1100 248 443*

255.vortex 1900 268 710* 1900 264 719*256.bzip2 1500 389 386* 1500 375 400*

300.twolf 3000 784 382* 3000 776 387*

SPECint_base2000 438

SPECint2000 442

http://www.spec.org/osg/cpu2000/results/res2000q3/cpu2000-20000718-00168.asc

Intel OR840(1 GHz

Pentium III processor)

8/13/2019 ACA UNit 1

21/29

21

Benchmarks

Performance Evaluation

For better or worse, benchmarks shape a field

Good products created when have:

Good benchmarks

Good ways to summarize performance

Given sales is a function in part of performance relative to

competition, investment in improving product as reported by

performance summary

If benchmarks/summary inadequate, then choose between

improving product for real programs vs. improving product to get

more sales;

Sales almost always wins!

Execution time is the measure of computer performance!

8/13/2019 ACA UNit 1

22/29

22

Benchmarks

Management would like to have one number.

Technical people want more:

1. They want to have evidence of reproducibilitythere should be enough

information so that you or someone else can repeat the experiment.

2. There should be consistency when doing the measurements multiple

times.

How to Summarize Performance

How would you report these results?

Computer A Computer B Computer C

Program P1 (secs) 1 10 20

Program P2 (secs) 1000 100 20

Total Time (secs) 1001 110 40

8/13/2019 ACA UNit 1

23/29

23

Quantitative Principles

of Computer Design

Make the common case fast.

Amdahls Law:Relates total speedup of a system to the speedup of some

portion of that system.

8/13/2019 ACA UNit 1

24/29

24

Amdahl's Law

Suppose that enhancement E accelerates a fraction F

of the task by a factor S, and the remainder of the

task is unaffected

Quantitative

Design

tEnhancemenWithoutePerformanc

tEnhancemenWithePerformanc

tEnhancemenWithTimeExecution

tEnhancemenWithoutTimeExecution

ESpeedup __

__

___

___

)(

Speedup due to enhancement E:

This fraction enhanced

Q

8/13/2019 ACA UNit 1

25/29

25

ExTimenew = ExTimeoldx (1 - Fractionenhanced) + Fractionenhanced

Speedupoverall =ExTime

old

ExTimenew

Speedupenhanced

=

1

(1 - Fractionenhanced) + Fractionenhanced

Speedupenhanced

Quantitative

Design

This fraction enhanced

ExTimeold ExTimenew

Amdahl's Law

8/13/2019 ACA UNit 1

26/29

26

Amdahl's Law

Floating point instructions improved to run 2X; but only

10% of actual instructions are FP

Speedupoverall =1

0.95= 1.053

ExTimenew= ExTimeoldx (0.9 + .1/2) = 0.95 x ExTimeold

Quantitative

Design

8/13/2019 ACA UNit 1

27/29

27

Quantitative

Design

Instruction Frequency

Invest Resources where time is Spent!

CPI = (CPU Time * Clock Rate) / Instruction Count

= Cycles / Instruction Count

n

iii ICPITimeCycleTimeCPU 1 **__

n

i

ii FCPICPI1

* whereCountnInstructio

Ii

iF_

Number of

instructions of

type I.

Cycles Per

Instruction

8/13/2019 ACA UNit 1

28/29

28

Quantitative

Design

Base Machine (Reg / Reg)

Op Freq Cycles CPI(i) (% Time)ALU 50% 1 .5 (33%)

Load 20% 2 .4 (27%)

Store 10% 2 .2 (13%)

Branch 20% 2 .4 (27%)

Total CPI 1.5

Suppose we have a machine where we can count the frequency with whichinstructions are executed. We also know how many cycles it takes for

each instruction type.

Cycles Per

Instruction

How do we get CPI(I)?

How do we get %time?

8/13/2019 ACA UNit 1

29/29

29

Quantitative

Design

Locality of

Reference

Programs access a relatively small portion of the address space at

any instant of time.

There are two different types of locality:

Temporal Locality(locality in time): If an item is referenced, it will

tend to be referenced again soon (loops, reuse, etc.)

Spatial Locality(locality in space/location): If an item is referenced,

items whose addresses are close by tend to be referenced soon

(straight line code, array access, etc.)