Over View of Microprocessor Systems

8/3/2019 Over View of Microprocessor Systems

http://slidepdf.com/reader/full/over-view-of-microprocessor-systems 1/45

NITW/ECE/LA 1

OVER VIEW OFMICROPROCESSOR

SYSTEMS

L.ANJANEYULU

Dept of ECE

N.I.T., Warangal

OVER VIEW OFMICROPROCESSOR

SYSTEMS

L.ANJANEYULU

Dept of ECE

N.I.T., Warangal



NITW/ECE/LA 2

ENIAC-on-a-ChipMoore School of Electrical Engineering, University of

Pennsylvania http://www.ee.upenn.edu/~jan/eniacproj.html



NITW/ECE/LA 3

Intel1950's: Shockley leaves Bell Labs to establish Shockley Labs inCalifornia. Some of the best young electronic engineers and solid-state physicists come to work with him. These include Robert Noyceand Gordon Moore.1969: Intel was a tiny start-up company in Santa Clara, headed by

Noyce and Moore .1970: Busicom placed an order with Intel for custom calculator chips.Intel had no experience of custom-chip design and sets outs to design ageneral-purpose solution .1971: Intel have problems translating architectures into working chipdesigns - the project runs late.

Faggin joins Intel and solves the problems in weeks.

The result is the Intel 4000 family (later renamed MCS-4,Microcomputer System 4-bit), comprising the 4001 (2k ROM), the4002 (320-bit RAM), the 4003 (10-bit I/O shift-register) and the 4004 ,a 4-bit CPU.



NITW/ECE/LA 4

Intel 4004Introduced in 1971, the Intel4004 "Computer-on-a-Chip" wasa 2300 transistor device capableof performing 60,000 operations

per Second.

It was the first-ever single-chipmicroprocessor and hadapproximately the same

performance as the 18,000vacuum tube ENIAC. The 4-bitIntel C4004 ran at a Clock Speedof 108 Kilo Hertz.



NITW/ECE/LA 5

The Intel 4004

FedericoFaggindesigned theIntel 4004

processor.His initialswere printedon thecircuit.



NITW/ECE/LA 6

Intel 4004 ± First Microcomputer

http://uk.geocities.com/magoos_universe/4004_main.htm



NITW/ECE/LA 7

The Busicom Calculator

The Busicom calculator used five Intel 4001¶s, two4002¶s, three 4003¶s andthe 4004 CPU

The original engineering prototype

of the Busicom desk-top printingcalculator, the world¶s firstcommercial product to use a

microprocessor.

http://www.computerhistory.org/exhibits/highlights/busicom.shtml



NITW/ECE/LA 8

Intel 8008

1972: Faggin begins work on an 8 -bit processor, theIntel 8 00 8 . The prototype has serious problems withelectrical charge leaking out of its memory circuits.

Device physics, circuit design and layout are important new skills. The 8008 chip layout is completely redesignedand the chip is released.There is a sudden surge in microprocessor interest.Intel's 8008 is well-received, but system designers wantincreased speed, easier interfacing, and more I/O andinstructions. The improved version, produced by Faggin,is the 8 08 0.Faggin leaves Intel to start his own company Zilog , who

later produce the Z 8 0.



NITW/ECE/LA 9

Federico Faggin : ZilogZilog produced the3.5MHz Zilog Z80 (a very

popular processor taught

in many universities)« and, later, a 16-bit Z8000.

Another great design butZilog struggled to providegood support, they were anew and inexperiencedcompany and had only afew hundred employees;at this time Intel had over 10 thousand.



NITW/ECE/LA 10

The Zilog Z80

The Z80 microprocessor is an 8 bitCPU with a 16 bit address bus capableof direct access of 64k of memoryspace.

It was based on the 8080; it has a largeinstruction set.

Programming features include an

accumulator and six eight bit registersthat can be paired as 3-16 bit registers.In addition to the general registers, astack-pointer, program-counter, andtwo index (memory pointers) registers

are provided.



NITW/ECE/LA 11

Early Microcontrollers1974: Motorola (originally car radio manufacturers) hadintroduced transistors in the 1950s and decided to make a late

but serious effort in the microprocessor market. Theyannounced their 8-bit 68 00 processor. Though bulky, andfraught with production problems, their 6800 had a gooddesign.1975: General Motors approach Motorola about a custom-builtderivative of the 6800. Motorola's long experience withautomobile manufacturers pays off and Ford follow GM's lead.

1976: Intel introduce an 8-bit microcontroller, the MCS-48.They ship 251,000 in this year.

1980: Intel introduce the 8 051 , an 8-bit microcontroller withon-board EPROM memory. They ship 22 million and 91million in 1983.



NITW/ECE/LA 12

The Intel 8086(1978)29,000 TransistorsClock Speeds: 5, 8 and 10 MHzApprox. 10 times the

performance of the 8080 Intel8086, 16 bit ³assembly-languagecompatible´ extension of the8080 architecture. 1978.

All registers 16 bits wide.Additional registers all havededicated uses.

Ex tended Accumulator architecture.IBM selects the 8088, an 8086with an 8 bit external bus, as the

processor for the IBM PC. Early1980



NITW/ECE/LA 13

Early Computers1979: Motorola also announce a 16-bit68000. Indisputably, the best

microprocessor on the market. It would beused in the Apple Macintosh launched in1984.

Intel look seriously at the competition(Motorola and Zilog) and implement'O peration CRUSH ' - a huge campaignwith a focused and trained work force

providing customer support, completesolutions and long-term product support.

CRUSH proves an excellent strategy andthe 8086 becomes the de facto standard.This success helps finance additions to their

product range, one of which is the buswidth reduced 8 088 , a 16-bit ( 8 -bit bus)microprocessor .

The early Apple Macintosh



NITW/ECE/LA 14

The IBM PC1981: IBM, having seen Apple's success recognise

a new personal computer market. They choose Intelover Motorola and Zilog (and their own proprietary

processors) because of Intel's long-term commitmentto the 8086 line.

IBM selects the Intel 8 088 for their PC, introduced inAugust.Intel bring out the 16-bit 80286 for the IBM PC AT

but it has weaknesses, most notably in virtual memorysupport. The newest 'killer' application software,Microsoft Windows , needs a more powerful

processor.



NITW/ECE/LA 15

Contemporary Microprocessors:

16/32-bit Processors(external 16-bit Bus, internal 32 BitStructure)

Motorola MC68010

National Semiconductor NS16032 Additional Functionality on the Chip

Direct Memory Access (DMA) (Intel 80186)Virtual memory management(MC68010, Intel 80286)Optional Coprocessor (Intel 8086/80286,NS16032)Extended Address Space



NITW/ECE/LA 16

32-bit ProcessorsCISC Processors

Motorola MC680x0Intel i386 / i486 / PentiumNational Semiconductor NS32x32Concept of a Processor FamilyBinary CompatibilityCompatible with 16 Bit Processors

RISC Processors

Advanced Micro Devices Am29000 (~1987)Sun Microsystems SPARCMIPS technologies MIPS R2000 / MIPS R3000

Microprocessor History



NITW/ECE/LA 17

Moore¶s Law

Dr. Gordon E. Mooreco-founded Intel in1968.

His observation thatnumber of transistorsdoubled every 2 years

became known as´ Moore ·s Law µ



NITW/ECE/LA 18

Pentium Evolution (1)8086

much more powerful16 bitinstruction cache, prefetch few instructions8088 (8 bit external bus) used in first IBM PC

80286(1982) :Added new instructions to support memory management.Added memory mapping and multilevel protection schemeAdded real addressing mode to support legacy 8086 code.

16 Mbyte memory addressable (up from 1Mb) Include:Segment limit checking,Read-only and execute-only segment options,Up to four privilege levels to protect operating system code (inseveral subdivisions, if desired) from application or user programs.

Hardware task switching and local descriptor tables allow the operatingsystem to protect application or user programs from each other.



NITW/ECE/LA 19

80386(1985) IA-32 architecture family .Support for multitaskingAdditional registers (segment pointers).All GP registers now 32 bits.Address space now 32 bits with several new addressingmodes. Provides logical address space for eachsoftware processAdded paging support under existing segmentedarchitecture. Supports:Segmented-memory modeland³ Flat ´ one-memory model

Almost a general purpose register machine.Intel386 Processor Includes 6 Parallel Stages. Bus interface u nit , Code prefetch u nit ,In s tr u ction decodeu nit , Exec u tion u nit , Segment u nit , Paging u nit



NITW/ECE/LA 20

Intel486 processor more parallel execution capability than Intel386instruction decode and execution units in five pipelined stages,

each stage (when needed) operates in parallel with the others on upto five instructions in different stages of execution.Each stage can do its work on one instruction in one clock, and sothe Intel486 processor can execute as rapidly as one instruction per clock cycle.

8-KByte on-chip fir s t level cache to increase the percent of instructionsthat could execute at the scalar rate of one per clock:Memory access instructions included if the operand was in the first-levelcache.

Integrated the x87 floating point unit onto the processor New pins, bits and instructions to support more complex and powerfulsystems

Second-level cache supportMultiprocessor support.Power management (for notebooks and laptops



NITW/ECE/LA 21

Speeding it upPipeliningOn board cache

On board L1 & L2 cacheBranch predictionData flow analysisSpeculative execution



NITW/ECE/LA 22

Pentium Evolution (2)

80486sophisticated powerful cache and instruction pipelining

built in maths co-processor

PentiumSuperscalar Multiple instructions executed in parallel

Pentium ProIncreased superscalar organization

Aggressive register renaming branch predictiondata flow analysisspeculative execution

CPU+ L1 Cache

L2 CacheCPU+ L1 Cache



NITW/ECE/LA 23

Pentium Evolution (3)Pentium II ( 1997)Pentium Pro + MMX (MultiMedia eXtensions)

- Data Bus (64bit), Address Bus (36bit)- L1 Cache: 32K Byte, L2 Cache: 512K Byte Cache- Processor Core Speed (450MHz - 233MHz)- System Bus (100MHz)

graphics, video & audio processingCeleron = Pentium II - L2Cache - Celeron A :

L2Cache(128KByte)Xeon = Pentium II + Graphic Accelerator + .. (Server CPU) (1998)Scalability : can be scaled to 2, 4, 8 or more,

and used for high-end server and workstations



NITW/ECE/LA 24

Pentium III(1999)

- Data Bus (64bit), Address Bus (36bit)- Processor Core Speed (1.1GHz - 450MHz)- System Bus (133MHz)- Cache Speed Upgrade (Advanced Transfer Cache)- 70 New Extended Instructions (SIMD)

70 new streaming SIMD extensions (SSE) :50 to improve floating-point performance

12 to improve multimedia processing8 to improve the efficiency of L1 cache

- Pentium III Xeon ProcessorAdditional floating point instructions for 3Dgraphics



NITW/ECE/LA 25

Pentium 4 ( 2000)

Further floating point and multimedia enhancements- Data Bus (64bit), Address Bus (36bit)- Processor Core Speed (2GHz - 3.2GHz)- System Bus (400MHz-800MHz)

- 800 MHz : Pentium 4 C-3.20 GHz, 3 GHz, 2.80 GHz, 2.60 GHz, 2.40 GHz

- 533 MHz : Pentium 4 B3.06 GHz, 2.80 GHz, 2.66 GHz, 2.53 GHz, 2.40 GHz, 2.26 GHz

- 400 MHz : Pentium 4 A2.60 GHz, 2.50 GHz, 2.40 GHz, 2.20 GHz, 2 GHz

-hyper-threading technologyItanium

64 bitSee Intel web pages for detailed information on processors



NITW/ECE/LA 26

Technological Development

420000002000Pentium 4

240000001999Pentium-III

75000001997Pentium-II

31000001993Pentium

1180000198980486

275000198580386

120000198280286

2900019788086500019748080

250019728008

225019714004

# of tran s is tor s Year Model



NITW/ECE/LA 27

Technological Development

1000

10000

100000

1000000

10000000

100000000

1 9 7 1

1 9 7 2

1 9 7 4

1 9 7 8

1 9 8 2

1 9 8 5

1 9 8 9

1 9 9 3

1 9 9 7

1 9 9 9

2 0 0 0

Y ear

# o

f t r a n s

i s t o r s

4004

8008

8080

8086

8028680386

80486 Penti u m

Penti u m IIPenti u m III

Penti u m 4



NITW/ECE/LA 28

Performance

4004 8 00 8 8 08 0 8 08 6 8 088

Introduced 1971 1972 1974 1978 1979

Clock Speeds

108 KHz 108 KHz 2 MHz 5 MHz,8MHz,10MHz

5 MHz,8MHz

Bus Width 4 bits 8 bits 8 bits 16 bits 8 bits

Number of

Transistors

2300 3500 6000 29,000 29,000

AddressableMemory

640 bytes 16 KBytes 64 KBytes 1 MB 1 MB

VirtualMemory

-- -- -- -- --

1970s Processors:



NITW/ECE/LA 29

Performance

8 02 8 6 3 8 6TM DX 3 8 6TM SX 4 8 6TM DXCPU

Introduced 1982 1985 1988 1989

Clock Speeds 6 MHz ±

12.5 MHz 16 MHz-33MHz 16 MHz-33MHz 25 MHz- 50MHz

Bus Width 16 bits 32 bits 16 bits 32 bits

Number of Transistors 134,000 275,000 275,000 1.2 million

AddressableMemory

16 MB 4 GB 4GB 4GB

Virtual

Memory

1 GB 64 TB 64 TB 64 TB

19 8 0s Processors:



NITW/ECE/LA 30

Performance

48 6TM SX Pentium Pentium Pentium II

Introduced 1991 1993 1995 1997

Clock Speeds

16 MHz-133MHz

60 MHz ± 166 MHz

150 MHz-200MHz

200 MHz-300MHz

Bus Width 32 bits 32 bits 64 bits 64 bits

Number of Transistors

1.185 million 3.1 million 5.5 million 7.5 million

AddressableMemory

4 GB 4 GB 64 GB 64 GB

VirtualMemory

64 TB 64TB 64 TB 64 TB

1990s Processors:



NITW/ECE/LA 31

Performance

Pentium III Pentium 4

Introduced 1999 2000

Clock Speeds

450 MHz 1.3-1.8 GHz

Bus Width 64 bits 64 bits

Number of Transistors

95 million 42 million

AddressableMemory

64 GB 64 GB

VirtualMemory

64 GB 64 TB

Recent Processors:



NITW/ECE/LA 32

Contemporary Microprocessor

64/32-bit ProcessorsSUN Microsystems SuperSPARCMotorola 88110IBM, Motorola PowerPC 601 (MPC601)

³ Modern ´ Processors64-bit StructureInternal Parallelism

Instruction pipeliningArithmetic Pipelining

Instruction and Data Caches Advanced Memory and PeripheralConnections



NITW/ECE/LA 33

Performance Mismatch

Processor speed increasedMemory capacity increased

Memory speed lags behind processor speed!!



NITW/ECE/LA 34

DRAM and Processor Characteristics



NITW/ECE/LA 35

Intel Itanium 2 (McKinley)

64bit Processor 221 million transistors!

How are they used?What will we do astransistor countscontinue to grow?Most of chip is used for memories, inst. decoding,dynamic scheduling«

Why is it done this way?How much more efficientcould it be if more of areawent to actual processing?



NITW/ECE/LA 36

Even More Recent Example

Runs 64-bitIA-64 ISA

Die: 3.74 cm 2

.13µ process410M transistors1.5GHz core1.3V logic130W power

consumption!

6.4GB/s busCost: $2,247-

$4,2269MB L3 cachelater this year«



NITW/ECE/LA 37

AMD Opteron (100 Million Transistors)



NITW/ECE/LA 38



NITW/ECE/LA 39

Cyrix IIIDeveloped by National Semiconductor 133 MHz Front Side Bus (although it supports 66 MHz, and 100 MHz FSB).256 KB integrated L2 cache along with a 64 KB integrated L1 cache.3dNow! SIMD instructions in a dual pipelined FPU.

As with the MII, the Cyrix III supports MMX.superscalar design featuring two seven stage pipelines allowing two processing streams to be processed simultaneously.two level translation buffer and a 512 entry branch target buffer.out-of-order execution through register renaming and data forwarding and

bypassing to resolve data dependencies between pipelines.Speculative execution after a predicted branch is also supported.15% to 20% cheaper than a comparable Celeron.Hope to capture 10% of market.Subject of a lot of legal action by Intel but VIA is still in business.



NITW/ECE/LA 40

First Implementation of Key Features: Montecito

Power Management/

FrequencyBoost

(Foxton)

1MB L2I1MB L2I

Dual-core

2x12MB L32x12MB L3cachescaches

withwithPellstonPellston

2 Way2 WayMultiMulti- -threadingthreading

Arbiter Arbiter

9 0 nm9 0 nm

1 .7 Billion1 .7 Billion

TransistorsTransistors

Key Processor FeaturesIntel¶s first dual-coreprocessor Intel¶s first processor

with >1 billion transistors24 MB L3 cacheMulti-threadingCompatible with existingItanium 2-based systems

Targeting H2¶2005

Sy s tem Bus

Core

L3 Cache

Core

L3 Cache

Sy s tem Bus

Core

L3 Cache

Core

L3 Cache

Core 1 Core 2

Mu ltiple core s , Mu ltiple thread s Mu ltiple core s , Mu ltiple thread s and L 3 Cache on ONE dieand L 3 Cache on ONE die

Mu ltiple core s , Mu ltiple thread s Mu ltiple core s , Mu ltiple thread s and L 3 Cache on ONE dieand L 3 Cache on ONE die



NITW/ECE/LA 41

Intel¶s Latest: The Pentium 4 2.4GHz

478 pin packaging



NITW/ECE/LA 42

Selecting a Microprocessor Issues

Technical: speed, power, size, costOther: development environment, prior expertise, licensing, etc.

Speed: how evaluate a processor¶s speed?Clock speed ± but instructions per cycle may differ Instructions per second ± but work per instr. may differ

Dhrystone: Synthetic benchmark, developed in 1984.Dhrystones/sec.

MIPS: 1 MIPS = 1757 Dhrystones per second (based onDigital¶s VAX 11/780). A.k.a. Dhrystone MIPS. Commonlyused today.

So, 750 MIPS = 750*1757 = 1,317,750 Dhrystones per second

SPEC: set of more realistic benchmarks, but oriented to desktopsEEMBC ± EDN Embedded Benchmark Consortium,www.eembc.org

Suites of benchmarks: automotive, consumer electronics,

networking, office automation, telecommunications



NITW/ECE/LA 43

Which has higher performance?Time to do the task (Execution Time)

execution time, response time, latencyTasks per day, hour, week, sec, ns. .. (Performance)throughput, bandwidth

Response time and throughput often are in oppositionResponse Time

Time to complete a task Throughput

Total amount of work done per timeExecution Time (CPU Time)

User CPU timeTime spent in the program

System CPU timeTime spent in OS

Elapsed TimeExecution Time + Time of I/O and time sharing



NITW/ECE/LA 44

Criteria of PerformanceExecution time seems to measure the power of theCPUElapsed time measures the performance of wholesystem including OS and I/OUser is interested in elapsed timeSales people are interested in the highest number of performance that can be quotedPerformance analysist is interested in bothexecution time and elapsed time

Performance evaluationPerformance evaluation



NITW/ECE/LA 45

Coffee Time!

Over View of Microprocessor Systems

Documents