204321 Computer Architecture

1
??????????????????????????

• ?????????? ??????????
• ??????????????????????????
• ??????????????????????

2
???????? ??????????????????????

• Baer
• the design on the integrated system which
  provide a useful to the programmer
• Hayes
• The study of the structure, behavior and design
  of computers
• ABD-Alla
• The design of the system specification at a
  general or subsystem level

3
???????? ??????????????????????

• Foster
• The art of designing a machine that will be a
  pleasure to work with
• Hennessy and Patterson
• The interface between the hardware and the
  lowest level software

4
???????? ??????????????????????

• Common themes
• Design
• Art
• System
• Tools
• Interface

• ??????
• ?????????????? ??????????????
• ?????
• ??????????????????????????????????????????????????
  ????????
• ?????????????????
• ????????????????????????

5
Architecture / Organization

• ????????
• Architecture ????????????????
• Organization ???????????????????????????????????
  
• ????????
• AMD / Pentium
• Branded PC ??? Pantip PC

6
?????????????????????
Input/Output and Storage
Disks, WORM, Tape
RAID
Emerging Technologies Interleaving Bus protocols
DRAM
Coherence, Bandwidth, Latency
Memory Hierarchy
L2 Cache
L1 Cache
Addressing, Protection, Exception Handling
VLSI
Instruction Set Architecture
Pipelining, Hazard Resolution, Superscalar,
Reordering, Prediction, Speculation, Vector, DSP
Pipelining and Instruction Level Parallelism
7
???????????????????????????????????????

• Instruction Set Architecture
• Pipelining, Hazard Resolution, Superscalar,
  Reordering, Prediction, Speculation, Vector, DSP
• Memory
• Memory Hierarchy, Memory Devices, Memory
  Management, Cache, Storages Disks, Tapes, RAID
• Input/Output
• Bus, I/O Interface, I/O control

8
?????????????????

• Mechanic
• Electronic
• ?????? 1 Vacuum tube
• ?????? 2 Transistor
• ?????? 3 LSI
• ?????? 4 VLSI
• ?????? 5 ???

9
???????/??????????

• Mechanical engine
• Charles Babbage (1822) Analytic Engine
• George Boole Mathematical analysis of logic
• ??????????????????????????????????????????????????
  ??
• ???, ???, ??????????????????

10
?????? 1 Vacuum tube

• ENIAC
• Eckert / Mauchly _at_ U. of Penn.
• ???????????????????
• 18000 tubes / 70,000 resistors / 10000 capacitors
  / 6000 switches / 915 meters / 140 kW
• ?????? 10
• ??????????? ?????????????
• 5000 ??????/?????? (add ??????????)

11
?????? 1 Vacuum tube

• IAS (Institute for Advanced Studies)
• Von Neumann and Goldstine
• ??????????? ENIAC
• ?????? von Neumann architecture
• ????????????????????????????????????????????
• Sequential execution
• Register sets (MBR, MAR, IR, IBR, PC, AC)

12
Structure of von Nuemann machine
Arithmetic and Logic Unit
Input Output Equipment
Main Memory
Program Control Unit
13
Structure of IAS - detail
Central Processing Unit
Arithmetic and Logic Unit
MQ
Accumulator
Arithmetic Logic Circuits
MBR
Input Output Equipment
Instructions Data
Main Memory
PC
IBR
MAR
IR
Control Circuits
Address
Program Control Unit
14
Commercial Computers ???????? 1

• 1947 - Eckert-Mauchly Computer Corporation
• UNIVAC I (Universal Automatic Computer)
• US Bureau of Census 1950 calculations
• Became part of Sperry-Rand Corporation
• Late 1950s - UNIVAC II
• Faster
• More memory

15
?????? 2

• Transistors
• ????????
• ?????????????
• ???????
• ????????
• Floating points operations
• Fortran
• Magnetic core memory
• DEC PDP-1
• IBM 7000 series

16
?????? 3

• Integrated Circuits
• Microprogramming
• Structured programming
• Semiconductor memory
• Mainframe IBM 360/370
• Mini PDP 11

17
IBM 360 series

• Replaced ( not compatible with) 7000 series
• First planned family of computers
• Similar or identical instruction sets
• Similar or identical O/S
• Increasing
• speed
• number of I/O ports (i.e. more terminals)
• memory size
• cost

18
DEC PDP-8

• Minicomputer ?????????? (after miniskirt!)
• ????????????????????????
• ???????????????????????????
• 16,000
• 100k for IBM 360
• Embedded applications OEM
• BUS STRUCTURE

19
DEC - PDP-8 Bus Structure
I/O Module
Main Memory
I/O Module
Console Controller
CPU
OMNIBUS
20
Semiconductor Memory

• Fairchild
• Size of a single core
• i.e. 1 bit of magnetic core storage
• 256 ???
• Non-destructive read
• Much faster than core
• ??????????????? 2 ?????????

21
?????? 4

• LSI / VLSI
• Packaged software
• OOP
• AI (Functional programming)
• Singleboard PC
• IBM PC / Cray XMP / IBM 3090

22
Intel

• 4004
• Microprocessor ??????
• ?????????????????????????????
• 4 ???
• ??????????????????????
• 8008
• 8 bit
• ??????????????????????
• 8080
• 8 ???
• ???????????????????????

23
?????? 5

• VLSI
• Internet
• Network computing Client-server, agent,
  distributed
• Parallel computers SMP, clusters
• Parallel programming

24
??????????????????????????????????

• On the Net
• In a box
• Inside CPU
• Inside Control Unit of CPU

25
On the Net

• ???????????????????????????????????? 1 ???
• Multicomputer / Parallel systems
• Clusters
• Client-server
• ???????????????

26
In a box

• ????????????????????????????????????
• ????????????????????????
• CPU
• Main memory
• I/O
• System interconnection

27
Structure In a box
Computer
Peripherals
Central Processing Unit
Main Memory
Computer
Systems Interconnection
Input Output
Communication lines
28
Inside CPU

• ?????????????????????????????? CPU
• ????????????????????????
• Control Unit (CU)
• Registers
• Arithmetic and Logic Unit (ALU)
• Inside-CPU interconnection

29
Structure - The CPU
CPU
Arithmetic and Login Unit
Computer
Registers
I/O
CPU
System Bus
Internal CPU Interconnection
Memory
Control Unit
30
Inside Control Unit of CPU

• Specific-task Registers
• Gate-level computing
• Microprogram / hardwire control

31
Structure - The Control Unit
Control Unit
Arithmetic and Login Unit
CPU
Registers
ALU
Control Unit
Internal Bus
Internal CPU Interconnection
Registers
Control Unit
32
????????????????????????????????

• ?????????????????????????????????????????
• ????????????
• ???????? Human genome, ??????????
• ????????????? ????, ???????, ???????????,
  ??????
• ??????? ??????????, ???????, ????????, ?????
• ?????????????????? ?????????????????????
• ????? ??????????
• ??????????? ??????????????????????

33
Moores Law

• Gordon Moore ?????????????????? Intel
• ??? ???? ?????????
• ????????????????
• ?????????????????????????? 2 ??????? 18 ?????
• CPU ?????????? 2 ??????? 18 ?????
• ???????????
• ?????????????????? (?????????????????????????)

34
??????????????????????????????????????????????
?????????????????
Alpha 21264 15 million Pentium Pro 5.5
million PowerPC 620 6.9 million Alpha 21164 9.3
million Sparc Ultra 5.2 million
Moores Law

• ??????????? CMOS
• Die size 2X ??? 3 ??
• Line width ????????????????? 7 ??

35
??????????????????????????? (Single Chip DRAM)
?? ????(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
36
????????????????????

• ?????? ????????
• Logic 2x ??? 3 ?? 2x ??? 3 ??
• DRAM 4x ??? 3 ?? 2x ??? 10 ??
• Disk 4x ??? 3 ?? 2x ??? 10 ??

37
??????????????????????????

• 2523 ????????????????????????????? in ?proc
• 2538 ?????2 ???? 2532 ?????????????????????????
  ???

38
???????????????????????????????
1000
Supercomputers
100
Mainframes
???????????
10
Minicomputers
Microprocessors
1
0.1
1965
1970
1975
1980
1985
1990
1995
2000
??
39
?????????????????????????????
??????? ?????????
40
?????????????????????????????
??????????????????? ???????????????
Benchmarks
??????? ?????????
????????????????? ?????????????
41
?????????????????????????????
??????????????????? ???????????????
??????????????????? ??????????????????
Benchmarks
??????? ?????????
????????????????? ?????????????
????????????????? ?????????????
Workloads
42
??????????????????

• Benchmarks, Traces
• ?????????
• ????, delay, ???????????????????????????,
  ???????????????
• ????????????????? (????????????????????????)
• ????? ISA, ???, ????
• ???????????
• ????????????
• 90/10 ???????????? , 2/1 ??????
• ?? ?????????
• ??????????, etc.

43
Benchmarks

• SPEC
• System Performance Evaluation Cooperative
• ????????????????????????
• MIPS
• ???????????? Million instructions per second
• ????????? Meaningless instructions per
  second
• ?????????
• ?????????????????????????
• ?????????? 4 ???????????? 6 ???
• ?????????????????
• MFLOPS
• Million Floating-point operations per seconds
• ??????????? FP ?????????????????

44
??????????????????
??????????
?????????
DC to Paris
?????????
Throughput (pmph)
?????? 747
610 mph
6.5 hours
470
286,700
????????
1350 mph
3 hours
132
178,200

• ????
• Execution time (ExTime), response time, latency
• ??????????? (???????????????)
• Throughput, bandwidth

45
??????????????????

• ???????????????????????????????????
• ????????
• ????
• ??? ????????? / ????????????????????????
• Compatibility
• Software ?????????????????
• ?????????????

46
????????????????????? ??

• Metrics
• Execution time
• ??????????????????
• Response time
• ????????????????????????
• Throughput
• ??????????????????????????????
• Speed up
• ???????????????? execution time
  ??????????????????????
• ?????????
• ????????????????????????????
• ?????????????????????????? 2 ???
• ?? response time, ????? Throughput

47
Amdahl's Law

• ExTime w/o E Performance w/
  E
• Speedup(E) -------------
  -------------------
• ExTime w/ E Performance w/o
  E
• ????????????????????????????? ?????????????
• ??????
• ??? Execution Time ?????? Response time
• ?????

2
2
2
2
1
2
48
Amdahls Law
ExTimenew ExTimeold x (1 - Fractionenhanced)
Fractionenhanced
Speedupenhanced
1
ExTimeold ExTimenew
Speedupoverall

(1 - Fractionenhanced) Fractionenhanced
Speedupenhanced
49
Pop quiz

• ?????? FP ???????? 2X ?????????? 10
  ????????????????????????? FP

ExTimenew
Speedupoverall
50
?????

• ?????? FP ???????? 2X ?????????? 10
  ????????????????????????? FP

ExTimenew ExTimeold x (0.9 .1/2) 0.95 x
ExTimeold
1


1.053
Speedupoverall
0.95
51
?????????????????????????????????(Cycles per
instruction)

• Average Cycles per instruction
• CPI (CPU time Clock Rate) / Instruction Count
  
• Cycles / Instruction Count
• ?????????????? n ??????
• CPU time Cycle time ?i (CPIiIi)

52
???????? ??????? CPI

• Base Machine (Register / Register)
• Operation Freq Cycles CPI(i) ( Time)
• ALU 50 1 0.5 (33)
• Load 20 2 0.4 (27)
• Store 10 2 0.2 (13)
• Branch 20 2 0.4 (27)
• 1.5

Typical Mix
53
Pop quiz 2

• ????????????????? throughput / ?? response time
• ????? clock
• ?????????? processor
• ?????????????????????
• ?????????????????

54
????

• ?????????
• Capacity Speed
• 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
• 6 ???????? 16X CPU 2X DRAM size
• ???? Execution time, Response time.
• ??? Throughput, Bandwidth
• Speedup X ???????? Y n ???? ???????????
• ExTime(Y) Performance(X)
• --------- --------------
• ExTime(X) Performance(Y)

55
????

• Amdahls Law
• CPI Law
• ?????? Execution time ????????????????????????????
  ???????????????????????
• ???????????????????????? ????????????????????????
  ?????????????