Presentation on Intel product (processors) Presented to: Sir Imran Hanif Presented by: Muhammad Arshad Bs(cs)4 th Roll#06-36

Developmental History of Main-Line Intel CPUs

Intel 4004 specs Introduced in November of 1971 Was the first microprocessor 108 KHz clock speed Four bit adder A sixty-four bit index register An address incriminator And an eight bit instruction register and decoder

Intel 8086/8088 Introduced on June 8, 1978 First commercially successful 16 bit processor Too expensive for early computers, so the 8088 was developed with an 8 bit bus 8088 and 8086 ensured success of Intel corporation, because they were they start of the *86 family that was the basis for every IBM compatible PC

8086 specifications clock speeds of 5 MHz, 8MHz, and 10 MHz The 8086 could address 1 Meg of memory, while the 4004 could only address 640 bytes the 8086 had fourteen registers –general registers, segment registers, index registers, status and control register 16 bit processing power

Intel High Integration 16-bit Microprocessor fifteen to twenty of the most common microprocessor system components, providing twice the performance of the 8086 while operating at the same clock speed strategic flaw in the –released in early 1981 Intel believed that all computer manufactures would follow IBM’s lead and incorporate a built in DMA and timer chip

80186 Specifications ten more instructions to the op code set of the 8086 two different modes for the to run in –Enhanced –Standard

Intel Released of February of shipped at slightly faster clock speeds than the introduced protected mode memory addressing –Allows programmers to access memory above 1 meg –Virtual memory becomes a possiblility up to 1 Gig

Intel 80386SX/DX First truly mass market processor with 32-bit addressing. Able to address 4 gigabytes of RAM, and 64 terabytes of virtual memory. Delivered performance ranging from 5 MIPS in the first 16 MHz model to 11.4 MIPS in the fastest 33Mhz model Offered real mode, protected mode, and a ability to run many “virtual” real mode sessions simultaneously 275,000 Transistors with a 1-micron process.

Intel 80486SX/DX 1.2 Million Transistors with a 1 micron process Performance ranges from 20 MIPS in the 25 MHz DX model to 70.7 MIPS in the 100 MHz DX4 model DX model contained math co-processor and 8K L1 Cache Could address 4 gigabytes of RAM and 64 terabytes of virtual memory. DX2 version allowed the core of the processor to run at twice the bus speed. DX4 version allowed the core of the processor to run at three times the bus speed.

Intel Pentium Processor Name was chosen to thwart companies like AMD, Cyrix, etc. 3.1 Million Transistors using.8 micron process 4 GB RAM/64 TB Virtual Memory 64-Bit Data Bus Superscalar Design – had hardware for simultaneous instruction execution 66 MHz bus speed Double the amount of L1 cache: an 8 KB instruction cache and an 8 KB data cache More advanced math co-processor

Pentium error – look-up tables MMX – multimedia extensions Pentium MMX ran at a core voltage of 2.8v instead of the normal Pentium’s core voltage of 3.5v Pentium MMX done with.35 micron process Intel Pentium Processor

Intel Pentium Pro Originally called “P6” L2 cache was available in 256 KB, 512 KB, 1 MB, and 2 MB versions, and ran at full clock speed Clock speeds ranged from 166 to 200 MHz, with bus speeds of 60 and 66 MHz Ran on Socket 8 connector, which was the only socket the Pentium Pro supported Support for up to 4 processors in parallel Ran much better in NT Laid groundwork for future Intel processors.

The Pentium II processor Released May 1997, the second P6 architecture chip from Intel Used new socket called “Slot 1” Originally ran at 233 and 266 MHz The creation of a new motherboard made just for the PII, the i440LX, allowed for greater processing speeds In October 1997 the 300 MHz processor was produced and the 333 MHz, called the “Deschutes,” a few months later Deschutes was the first chip to implement.25 micron technology, previously.35 micron

Pentium II (Can't) Next Came the 350 and 400 MHz processors The first processors to introduce a 100 MHz front side bus April 1998 another motherboard was released, which led to the last Pentium II chip Running at 450 MHz with 512 KB of cache memory running at one half clock speed

Pentium II (Can't) ProcessorClock SpeedBus Speed Clock Multiplier PII MHz66 MHz3.5x PII MHz66 MHz4x PII MHz66 MHz4.5x PII MHz66 MHz5x PII MHz100 MHz3.5x PII MHz100 MHz4x PII MHz100 MHz4.5x

The Celeron processor In an attempt to regain the loses in the sub $1000 computer market, the Celeron processor was produced in early 1998 Based heavily on the Pentium II, except was without the level 2 cache and plastic casing The removal of the casing allowed for easier cooling, but having no level 2 cache SEVERLY reduced speed. Originally ran at 266 and 300 MHz Finally in August 1998 the Celeron A was introduced It had only a quarter the level 2 cache (128 MHz) of the Pentium II, but was implemented entirely different

The Pentium II Xeon processor June 29, 1998 the Pentium II Xeon processor was released to the public The processor was geared solely toward the high- end workstation and server market. It was not for home or personal use since servers often require a chip more powerful than 512 MB The Xeon was also built on the Deschutes core The level 2 cache was made with different chips called “CSRAM” (Custom Standard RAM), which was made specially by Intel Required extreme cooling so it had a very large casing

Xeon (Can't) The Xeon had some other special features -Can support up to 8 CPUs in one system -The cache was limited at 64 GB, done through 36-bit address bus -Runs a 100 MHz front side bus with a maximum band width of 800 MB/sec -Contains a built in thermal sensor -Used a new kind of ROM called PIROM which stantds for processor information ROM -This made the processor much more flexible to specifications

The Pentium III processor The Pentium III was to the Pentium II what the Pentium MMX was to the original Pentium Released in February 1999 Equipped with 70 new instructions to improve 3D and multi- media applications Main difference between the PIII and the other Pentiums was that it implemented the SSE instructions, which accelerates floating-point instructions over the display adapter

The Pentium IV processor The latest processor produced by Intel and also the fastest, from 1.6 to 2.4 GB Implements the Intel® NetBurst™ architecture –A 400 MHz system bus brought achieved through a physical signaling scheme of quad pumping the data transfers over a 100-MHz clocked system bus and a buffering scheme –Uses SSE2 which extends the old SSE with 144 new instructions, including 128 bit integer and floating point calculations, and an added data movement register was also added which improves both floating point and multimedia applications

Intel Processors In The Enterprise Business Development Intel GmbH, Munich

Outline What is a CPU? Current Technologies CPU, memory and Motherboards Concentrate on x86 architectures –INTEL –AMD Longer view Parallel CPUs Overview of Other vendors –SGI, SUN, COMPAQ, HP, IBM

What is a CPU? (1) A CPU is comprised of –Clock to 3GHz The controls how often an instruction can be performed –Integer Units 1 to N Used to perform integer maths –Floating point Units 1 to N 32/64bit arithmetic –Memory Cache L1, L2 and L3 Instruction and data Caches L1 Typically 8-64k L k L3 as large as 8MB –Memory Bus Speed –bus speed 100Mhz or higher Width –What is transferred per cycle Typically 64bit

What is a CPU? (2) –Memory Architecture SDRAM –One fetch per bus cycle DDR –Two fetches per bus Cycle RDRAM –Special Units SSE single precision (32 bits ) SIMD units SSE2 Double precision (64 bits) SIMD units –SIMD – single instruction multiple data Eg A(1:100)=A(1:100)*.08 Ie multiply each array element by 0.08.

Current Technologies - Intel (1) Intel Offerings Celeron and PIII –SSE, single precision SIMD units –BLAS libraries very fast (using ATLAS 700Mflops DGEMM (70% of peak) for 1GHz PIII (256k L2) –PIII dropped –Celeron moved to PIV core as of 1.7GHz –See for latest numbers

Current Technologies - Intel (2) Intel Offerings PIV, Xeon, Itanium (IA64) –PIV –BLAS libraries very fast (using ATLAS) SSEII - double precision SIMD 2.8Gflops for 2.2GHz P4 (using SSEII) –Xeon PIV core with SMT (symmetric multithreading) –Itanium 2 EPIC 3.5 Gflops DGEMM for 1000MHz Itanium 2 Price of itanium prohibitively expensive

Current Technologies - AMD (1) AMD Offerings Duron, AthlonXP and AthlonXP –Duron, (phased out?) AthlonXP for single CPU –AthlonMP required for Dual SMP –SSE, single precision SIMD units –BLAS libraries very fast (using ATLAS) 2.4 Glops DGEMM (75% of peak) 1.6GHz AthlonMP

Current Technologies - Motherboards PIII, PIV and AthlonMP available in Dual form –Both Xeons and AlthonMP cost more –a 1.8GHz AthlonMP costs ~1.5x 1.8GHz AthlonXP – a 2.4GHz Xeon costs ~1.5x 2.4GHz PIV –A 2.4GHz Xeon costs 1.3x 1.8GHz AthlonXP –64bit/66MHz PCI for both Motherboard costs –AthlonMP Tyan S2462UVM (SCSI, PCI, 100Mbit) –~x2 Xeon supermicro P4DP6 (SCSI, PCI-X,100Mbit) –1U rack mount routine –Blades becoming available for even higher density –PCI-X 64bit/133MHz Very interesting for high speed interconnects PIV Quad motherboards –Expensive –Limited memory bandwidth - bus based

CPUs on the horizon Intel (1) Very difficult predicting the future Intel. Celeron –1.7GHz, 1.9GHz, 2GHz (128 L2) Q3 and Q4 (P4 core) –single CPU only ? PIII –1.4GHz probably the last CPU? PIV/Xeon –3.06 GHz (512k L2) 533MHz bus (4*133) November 2002 –3.2 GHz (512k L2) 533MHz bus (4*133) Q XeonMP –high-end –256k L2, 1MB L3 1.6GHz –2GHz systems with 4 or more process. PIV Prescott – crystal ball gazing –3.2GHz, 4.0GHz, 1MB L2, 666 MHz bus Q3 2003, Q4 2003

CPUs on the horizon Intel (2) IA-64 –Compiler choice critical –32bit x86 code supported but how fast? McKinley Itanium 2 –1GHz 1.5MB-3MB L3 cache –400MHz bus (cf itanium 266MHz) –very expensive 10 time the cost of PIV? Madison –1.2/1.6 GHz > 3MB L3 cache 2H 2003

CPUs on the horizon AMD (1) AMD Duron silently dropped? AMD AthlonXP and MP lines – MHz FSB, 256k L2 cache –3000 and beyond 333MHz FSB, 512k L2 cache “Barton” H1 2003

CPUs on the horizon AMD(2) AMD HAMMER Series 4th Q st 2003 –64bit x86 CPU with 32bit x86 native –SSE and SSEII SIMD units –AMD Chipset (Hyper transport) PCI - X (133MHz)

Parallel CPUs (1) Parallel high end CPUs Itanium 2 –DMH (DDR memory Hub) Good memory bandwidth (6.4GB/s) –Poor scalability, all shared! –

Parallel CPUs (2) Parallel high end CPUs Hammer/Opteron –5.4GB/s memory bandwidth –Non shared therefore scales well –Ideal for memory intensive calcs Cc-numa problems for Linux? 2 way 4 way

Other vendors There are still other vendors in the market place SGI, SUN, COMPAQ, HP, IBM SGI –R14k 600MHz, Specint 483, Specfp 499 Not very fast but large scale cc-numa SMP systems 1024 procs Moving to IA-64 SUN –Ultra sparc III (1050MHz, Specint 537, Specfp 701 Speed OK but medium sized SMP systems COMPAQ/HP –Alphaserver systems - EV GHz, Specint 928, specfp 1327 fast cpu and systems cpu SMP Moving to IA-64 with HP HP –750MHz PA-8600, Specint 569, Specfp 526 –Heavily involved with IA-64 –1GHz Itaniumn 2 Specint 807, specfp 1356 IBM –Power4 Specint 804, Specfp 1202

Performance 1980 Time % increase per year from semiconductor technology 60-70% target perf increase Evolution of Microprocessor Technology 486/Pentium® processor RISC technologies 2 instructions / cycle Itanium™ processor EPIC: Even greater instructions / cycle 8086/286 CISC.3 ins / cycle bit architecture 1 instruction / cycle

Software on Intel® Xeon™ processor family More bandwidth, throughput and thread-level parallelism Better scalability for larger server workloads Outstanding platform performance Scale up to 4-way then scaling out in clusters of 4-way Optimum price / performance

Intel® Itanium™ Processor Family

SC What’s New? Redundant cooling –Including power supply fans All chassis fans now hot-swappable –Easy tool-free access and replacement of HS Fans –Two fans at rear, three inside Improved serviceability –Color coding of service “touch points” Green for hot swappable components, blue for service access points –Tool-less PCI card retention –Improved labeling Power and thermal support for up to two Intel ® Xeon™ processors

Centers Worldwide Dupont ® Tokyo Shanghai Moscow Santa Clara New Mexico London India Sweden Chantilly Third party trademarks and brands are the property of their respective owners Portland FolsomMunich Atlanta BrazilAustraliaCupertinoBeijingArizona Micro Centers Grenoble

® Thank you!