EGRE 426
-
Upload
belle-bradshaw -
Category
Documents
-
view
30 -
download
0
description
Transcript of EGRE 426
![Page 1: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/1.jpg)
1
EGRE 426
Handout 1
8/22/08
![Page 2: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/2.jpg)
2
Preliminary• EGRE 365 is a prerequisite for this class.• Class web page http://www.people.vcu.edu/~jhtucker/f08-egre426/index.html
• Syllabus Grades:– Quizzes (2) 40%– Homework 10%– Laboratory 10%– Design Project15% – Final Exam 25%
• Text book Computer Organization and Design 3rd edition, Designers Guide to VHDL.
![Page 3: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/3.jpg)
3
Computer Organization and Design
• This book and the slightly more advanced Computer Architecture a Quantitative Approach are the dominant computer architecture text books.
• Too big.– Chapter 1 – Read
– Chapter 2-7 – will be covered in detail skipping some material.
– Chapters 8-9 – Will depend on time.
– Supplement with additional material.
![Page 4: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/4.jpg)
4
Introduction
• We will learn not just how computers work, but gain an understanding into how and why computers evolved into the current generation.
• Computer have undergone rapid changes.– Increasing performance– Decreasing cost
![Page 5: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/5.jpg)
5
History
• Computers did not really begin until World War II.
• The widespread use of microprocessors began about 30 years ago.
• Personal computers were not taken seriously until the introduction of the IBM PC.
![Page 6: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/6.jpg)
6
Computers have resulted in the information revolution.
• Agricultural revolution.– Several Thousand years.
• Industrial revolution.– Several Hundred years.
• Information revolution.– A couple of decades.
![Page 7: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/7.jpg)
7
My History• I started at NASA in 1963 and worked in the
computer division.– At that time computers were very expensive and very
large.
– Programs were written on punched cards (one line per card) and fed into the computer.
– The state-of-the-art super computer that I worked with was capable of executing one million instructions per second, it cost several million dollars, and occupied the major portion of a large building.
![Page 8: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/8.jpg)
8
My History continued
• In the 60’s I did early work in interactive computing techniques.– Forerunner of what we do today, but not
practical when it required a dedicated super computer connected to a display console.
• Completed my PhD in 1974 and started working with microprocessors.
![Page 9: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/9.jpg)
9
Moore’s Law• One of the founders of Intel, Dr. Gordon Moore, observed
in 1965 that the number of transistors in an IC was doubling every year. He predicted that this would continue for a couple of decades and then slow to doubling every 18 months.
• This prediction has proved remarkably accurate. So much so that it has come to be expected, and Moore’s prediction has become known as Moore’s law.– It is worth noting that since Moore made his prediction the
consensus at any time has been that it would last for only about ten more years.
– Economics not technology may be what actually stops Moore’s law.
![Page 10: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/10.jpg)
10
![Page 11: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/11.jpg)
11
![Page 12: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/12.jpg)
12
![Page 13: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/13.jpg)
13
![Page 14: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/14.jpg)
14
![Page 15: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/15.jpg)
15
![Page 16: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/16.jpg)
16
![Page 17: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/17.jpg)
17
![Page 18: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/18.jpg)
18
![Page 19: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/19.jpg)
19
Definitions
Efficiency is the measurement of how close we come to achieving ideal speed up.
1
1
p
S
T
pT
E p
pp
![Page 20: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/20.jpg)
20
![Page 21: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/21.jpg)
21
Assumptions
• All operations take one unit of time.• All instructions and data are available when
needed.• i.e. We don’t have to wait for memory or
communication.• This is naïve and completely unrealistic, but
can be used to teach some fundamental truths.
![Page 22: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/22.jpg)
22
Conventional uniprocessor (SISD)
T A1*B1 A2*B2 A3*B3 A4*B4 … An*Bn
1
2
3
4
5
6
![Page 23: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/23.jpg)
23
Multiprocessor MIMD (unlimited processors)SIMD – same results.
T A1*B1 A2*B2 A3*B3 A4*B4 … An*Bn
1
2
3
4
5
6
![Page 24: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/24.jpg)
24
Multifunction computer (2 * units)
T A1*B1 A2*B2 A3*B3 A4*B4 … An*Bn
1
2
3
4
5
6
![Page 25: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/25.jpg)
25
SISD – single processor
T A1 A2 A3 A4 A5 An
1
2
3
4
5
6
n
n
ii AAAAP ...21
1
![Page 26: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/26.jpg)
26
MIMD or SIMD – with unlimited processors
T A1 A2 A3 A4 A5 An
1
2
3
4
5
6
n
n
ii AAAAP ...21
1
![Page 27: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/27.jpg)
27
Algorithm can effect speedup.SISD
T A (B C D + E) A B C D +A E1
2
3
4
5
6
AEABCDEBCDA )(
![Page 28: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/28.jpg)
28
Algorithm can effect speedup.MIMD
T A (B C D + E) A B C D +A E1
2
3
4
5
6
AEABCDEBCDA )(
![Page 29: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/29.jpg)
29
SISD Method A
T A0 + A1*X + A2*X*X + A3*X*X*X + A4*X*X*X*X
1
2
3
4
5
6
7
8
9
10
11
...432100
XXXXAXXXAXXAXAAXAPn
i
ii
![Page 30: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/30.jpg)
30
SISD Method B
T A0 + A1*X + A2*X*X + A3*X*X*X + A4*X*X*X*X
1
2
3
4
5
6
7
8
9
10
11
...432100
XXXXAXXXAXXAXAAXAPn
i
ii
![Page 31: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/31.jpg)
31
SISD Method C
T A0 + X ( A1 +X ( A2 + X ( A3 + X ( A4 + …))))
1
2
3
4
5
6
7
8
9
...))))((((
...
43210
432100
AXAXAXAXA
XXXXAXXXAXXAXAAXAPn
i
ii
![Page 32: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/32.jpg)
32
MIMD unlimited processors
T A0 + X ( A1 +X ( A2 + X ( A3 + X ( A4 + …))))
1
2
3
4
5
6
7
8
9
...))))((((
...
43210
432100
AXAXAXAXA
XXXXAXXXAXXAXAAXAPn
i
ii
![Page 33: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/33.jpg)
33
MIMD unlimited processors
T A0 + A1*X + A2*X*X + A3*X*X*X + A4*X*X*X*X
1
2
3
4
5
6
7
8
9
...432100
XXXXAXXXAXXAXAAXAPn
i
ii
![Page 34: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/34.jpg)
34
SIMD unlimited processors
T A0 + A1*X + A2*X*X + A3*X*X*X + A4*X*X*X*X
1
2
3
4
5
6
7
8
9
...432100
XXXXAXXXAXXAXAAXAPn
i
ii
![Page 35: EGRE 426](https://reader036.fdocuments.in/reader036/viewer/2022081603/5681369a550346895d9e35e5/html5/thumbnails/35.jpg)
35
1)(log222
22)(loglog)(log)( nnnKnnT
K K K( ) , ( ) , ( )0 0 1 1 2 1
K i K i i( ) ( ) 2 2 3
Where,
And,for i > 2
A X A A X A X A X A Xii
nn
i
n
0 1 22
33
11
0
1
Using an MIMD machine with unlimited processors, the time to compute
is given by