Eng Sem I #12 IC Fabrication & Microprocessors (1) PDF

Microprocessors (CPU’s)  A microprocessor is the main component of a microcomputer system and is also called as CPU (Central Processing Unit).  A microcomputer system consist of a minimum of memory (e.g. RAM & ROM) & I/O devices (Graphics Card, I/O devices, ect).  A microcomputer is a pro...

Microprocessors (CPU’s)  A microprocessor is the main component of a microcomputer system and is also called as CPU (Central Processing Unit).  A microcomputer system consist of a minimum of memory (e.g. RAM & ROM) & I/O devices (Graphics Card, I/O devices, ect).  A microcomputer is a programmable machine. Modern computers are electronic and digital. The two principal characteristics of a computer are:  It responds to a specific set of instructions in a well-defined manner.  It can execute a prerecorded list of instructions (a program) Byte Scale (Unit of Storage/Capacity) Definition: FLOP  Computer systems use floating-point numbers to represent extremely large numbers that would otherwise require many digits to record.  ICT professionals use the term "flops" to indicate how quickly computers can calculate these numbers.  The use of terms like "gigaflop" correspond to other terms like "gigabyte," which represents one billion individual bytes of data storage.  It’s important to note that in terms of processing speed and power, even the average device such as a laptop or desktop computer has already advanced beyond the capacity of a single gigaflop A gigaflop is equal to one billion floating- point operations per second. Floating-point operations are the calculations of floating- point numbers. Definition: FLOP Scientific Prefixes in Computing Name Unit Value kiloFLOPS kFLOPS 103 megaFLOPS MFLOPS 106 gigaFLOPS GFLOPS 109 teraFLOPS TFLOPS 1012 petaFLOPS PFLOPS 1015 exaFLOPS EFLOPS 1018 zettaFLOPS ZFLOPS 1021 yottaFLOPS YFLOPS 1024 Supercomputing Oak Ridge Facility Sumit Supercomputer Cost €200 million Scales in Computing Terascale: Refers to methods and processes for using supercomputers capable of performing at least 1 TFLOPS or storage systems capable of storing at least 1 TB Tera = 1 x 1012 (1024 GBytes) Petascale: Refers to methods and processes for using supercomputers capable of performing at least 1 PFLOPS or storage systems capable of storing at least 1 PB Peta = 1 x 1015 (approx 1 million Gbytes) Exascale: Refers to methods and processes for using supercomputers capable of performing at least 1 EFLOPS or storage systems capable of storing at least 1 EB Comparisons – Computing Power PlayStation 4 1.843 TFLOPS 2013 Xbox One S 1.4 TFLOPS 2016 PlayStation 4 4.2 TFLOPS 2016 Pro Nintendo 1 TFLOPS 2017 Switch Xbox One X Laptop/Desktop: 6 TFLOPS 2017 Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz 45.93 GFLOPS As of June 2018, the fastest machine was the Summit Supercomputer: IBM PowerPC and Nvidia machine at Oak Ridge. It hit 148.6 petaflops (148.6 x 10^15) on 2.2M cores Equivalent of about 63 gigaflops per core (CPU). It’s theoretical peak is 200 petaflops (200 x 1015 FLOPS) Works with 10PetaBytes of RAM Memory (=10 Million GBytes of RAM)  Block diagrams (Microcomputer & Microprocessor) Block diagram of a basic computer system Address bus ROM RAM I/O I/O CPU interface devices Data bus Control bus Microprocess or 9 Microcomputerr Hardware All general-purpose computers require the following hardware components:  Memory: Enables a computer to store data and programs.  Mass storage device: Allows a computer to permanently retain large amounts of data. Common mass storage devices include disk drives and tape drives.  Input device: Usually a keyboard and mouse are the input device through which data and instructions enter a computer.  Output device: A display screen, printer, or other device that lets you see what the computer has accomplished.  Central processing unit (CPU): The heart of the computer, this is the component that actually executes instructions.  Software The programs and data stored in a microcomputer is called as software. Programs can be written in low level languages or high level languages. A low level language can be binary language or assembly language.  A CPU recognizes only binary language which is called as machine language.  Assembly language instructions contain alphabets and/or numeric characters. To run assembly language programs a converter called as assembler is required. High level languages are more user friendly and contain simple words of English language. To run high level programs, converters such as compilers or interpreters are required. Input & Output Devices Input devices are used to input electrical or physical information in a microcomputer system in digital form.  In embedded applications, commonly used input devices are simple switches and sensors.  In general purpose microcomputers, input devices can be scanners, keyboard, mouse, network card etc. Output devices are used to display or perform required operation.  In embedded applications (CPU in a Washing Machine, for example) commonly used output devices are LED display units, LCD display units, stepper motors etc.  In general purpose computers output devices are mainly LCD screens, LED screens, Printers, network card etc. Memories Memory in a microcomputer system is used to store data and programs temporarily or permanently. The memories of primary concern for the CPU are only RAM & ROM which are called as primary memory or main memory. The CPU, at any one time, can only communicates with RAM & ROM. Other than primary memories, there are also secondary memories which are used for mass storage of data and programs and are transferred to the primary memory when required to be executed by the CPU. Examples of secondary memories are Hard Disks, Memory Classification Internal structure and basic operation of microprocessor Address bus ALU Register Section Data bus Control and timing section Control bus Block diagram of a microprocessor 15 Memory Mappin g for I/O Devices Memory Mapping In the following table, the increased maximum resources of computers that are based on 64-bit versions of Windows and the 64-bit Intel processor are compared with existing 32-bit resource maximums. Architectural 64-bit Windows 32-bit Windows component Virtual memory 16 terabytes 4 GB Paging file size 256 terabytes 16 terabytes Hyperspace 8 GB 4 MB Paged pool 128 GB 470 MB Non-paged pool 128 GB 256 MB System cache 1 terabyte 1 GB 17 Explanation: Virtual Memory This is a method of extending the available physical memory on a computer. In a virtual memory system, the operating system creates a pagefile, or swapfile, and divides memory into units called pages. Recently referenced pages are located in physical memory, or RAM. If a page of memory is not referenced for a while, it is written to the pagefile. This is called "swapping" or "paging out" memory. If that piece of memory is then later referenced by a program, the operating system reads the memory page back from the pagefile into physical memory, also called "swapping" or "paging in" memory. The total amount of memory that is available to programs is the amount of physical memory in the computer in addition to the size of the pagefile. An important consideration in the short term is that even 32-bit applications will benefit from increased virtual memory address space when they are running in Windows x64 Editions. Other Explanations Paging file: This is a disk file that the computer uses to increase the amount of physical storage for virtual memory. Hyperspace: This is a special region that is used to map the process working set list and to temporarily map other physical pages for such operations as zeroing a page on the free list (when the zero list is empty and the zero page is needed), invalidating page table entries in other page tables (such as when a page is removed from the standby list), and in regards to process creation, setting up the address space of a new process. Paged pool: This is a region of virtual memory in system space that can be paged in and out of the working set of the system process. Paged pool is created during system initialization and is used by Kernel-mode components to allocate system memory. Uniproccessor systems have two paged pools, and multiprocessor systems have four. Having more than one paged pool reduces the frequency of system code blocking on simultaneous calls to pool routines. Non-paged pool: This is a memory pool that consists of ranges of system virtual addresses that are guaranteed to be resident in physical memory at all times and thus can be accessed from any address space without incurring paging input/output (I/O). Non-paged pool is created during system initialization and is used by Kernel-mode components to allocate system  16 bit = 65, 536 bytes (64 Kilobytes) Theoreti cal Limits 32 bit = 4, 294, 967, 295 bytes (4 Gigabytes) for Address 64 bit = 18, 446, 744, 073, Busses 709, 551, 616 (16 Exabytes) System (Bus) System BUS (Three Types) The Diagram of a Microprocessor (µP) System Clock Explanation of a Microprocessor (µP) System  The microprocessor is a  In the µP based system, the microprocessor semiconductor device (Integrated is the master and all other peripherals are Circuit) manufactured by the VLSI slaves. The master controls all the peripherals and initiates all operations. The (Very Large Scale Integration) work done by the processor can be technique. It includes the ALU, classified into the following three groups. register arrays and control circuit  Work done internal to the processor on a single chip.  Work done external to the processor  A system designed using a  Operations initiated by the slaves or microprocessor as its CPU is called peripherals. a microcomputer.  The work done internal to the processors  The Microprocessor based system are addition, subtraction, logical operations, (single board microcomputer) data transfer operations, etc.  The work done external to the processor are consists of microprocessor as CPU, reading/writing the memory and semiconductor memories like reading/writing the I/O devices or the EPROM and RAM, input device, peripherals. If the peripheral requires the output device and interfacing attention of the master then it can interrupt devices. the master and initiate an operation.  The memories, input device, output  The microprocessor is the master, which device and interfacing devices are controls all the activities of the system. To called peripherals. perform a specific job or task, the microprocessor has to execute a program  The popular input devices are stored in memory. The program consists of a keyboard and floppy disk and the set of instructions. It issues address and output devices are printer, LED/LCD control signals and fetches the instruction Explanation of a Microprocessor (µP) System BUSES: PERIPHERAL DEVICES:  The buses are group of lines  The EPROM memory is used to that carries data, address or store permanent programs and control signals. data.  The CPU Bus has multiplexed  The RAM memory is used to store lines, i.e., same line is used to temporary programs and data. carry different signals  The input device is used to enter  The CPU interface is provided to the program, data and to operate demultiplex, the multiplexed the system. lines, to generate chip select  The output device is used for signals and additional control examining the results. signals. Since the speed of I/O devices  The system bus has separate does not match with the speed of lines for each signal. microprocessor, an interface  All the slaves in the system are device is provided between connected to the same system system bus and I/O devices. bus. At any time instant Generally I/O devices are slow communication takes place devices. between the master and one of Microprocessor (µP) Clock  In general, the clock refers to a microchip that regulates the timing and speed of all computer functions.  In the clock chip is a crystal (piezoelectrical crystal like quartz) that vibrates at a specific frequency when electricity is applied.  The shortest time any computer is capable of performing is one clock, or one vibration of the clock chip.  The speed of a computer processor is measured in clock speed, for example,  1 MHz is one million cycles, or vibrations, a second. System BUS – In Real Life Motherboard A CPU Cache CPU L1 RAM Cache A CPU cache is a small Memory memory location within Registers the CPU itself used by On-chip the CPU of a computer to reduce the average time to access data from the main memory. A cache will store recent data. A CPU Cache Levels  When the processor needs to read from or write to a location in main memory, it first checks whether a copy of that data is in the cache.  If so, the processor immediately reads from or writes to the cache, which is much faster than reading from or writing to main memory  Cache’s are faster then accessing memory, but may not always “prefetch” or hold the data the CPU needs.  An issue with speed versus accuracy is the fundamental tradeoff between cache latency and hit rate.  Larger caches have better hit rates but longer latency.  To address this tradeoff, many computers use multiple levels of cache, with small fast caches backed up by larger, slower caches.  Multi-level caches generally operate by checking the fastest, level 1 (L1) cache first; if it hits, the processor proceeds at high speed. 4004  First microprocessor (1971)  For Busicom calculator  Characteristics  10 mm process  2300 transistors  400 – 800 kHz  4-bit word size  16-pin DIP package  Masks hand cut from Rubylith  Drawn with color pencils  1 metal, 1 poly (jumpers)  Diagonal connectors 80286  Virtual memory (1982)  IBM PC AT  Characteristics  1.5 mm process  134k transistors  6-12 MHz  16-bit word size  68-pin PGA  Regular datapaths and internal ROMs 80386  32-bit processor (1985)  Modern x86 ISA  Characteristics  1.5-1 mm process  275k transistors  16-33 MHz  32-bit word size  100-pin PGA  32-bit datapath, microcode ROM, synthesized control Case Study: Intel Processors Slide 33 Pentium  Superscalar (1993)  2 instructions per cycle  Separate 8KB I$ & D$  Characteristics  0.8-0.35 mm process  3.2M transistors  60-300 MHz  32-bit word size  296-pin PGA  Caches, datapath, FPU, control Pentium 4  Deep pipeline (2001)  Very fast clock  256-1024 KB L2$  Characteristics  180 – 90 nm process  42-125M transistors  1.4-3.4 GHz  32-bit word size  478-pin PGA  Units start to become invisible on this scale Case Study: Intel Processors Slide 35 Intel™ iCores  Generational Multi-Core CPU’s  Many more contact points  4 Physical CPU’s = 8 logical (HyperThreading)  Architecture 22 nm technology  20 MB Intel® Smart Cache  Intel® 64 architecture  Level 1 Cache: 16KB data cache  Level 2 Cache Smart Cache dividing up 20MB between 8 cores.  Up to 3.5 Ghz with Intel® Turbo Boost Technology  64GB of addressable memory Intel™ iCore 7i Summary 104 increase in transistor count, clock frequency over 30 years! Newer Intel™ CPU’s Growth in Processing Growth in Scale (Moore’s Law) Moore’s Law  Moore's Law is the  In subsequent years, the observation made in 1965 pace slowed down a bit, but by Gordon Moore, co- data density has doubled founder of Intel approximately every 18  It States that the number months of transistors per square  This is the current definition inch on integrated circuits of Moore's Law had doubled every year  Most experts, including since the integrated circuit Moore himself, expect was invented. Moore's Law to hold true  Moore predicted that this until 2020-2025 trend would continue for  The limitation which exists is the foreseeable future. that once transistors can be created as small as atomic particles, then there will be The Future of CPU’s? Non-silicon alternative – carbon nano-tubes molecular computing with organic molecules Optical Computers – light instead of electricity Quantum computing- computing with atoms and their component parts Carbon Nano Tubes Faster Dissipate Heat Better Better Energy Efficiency Any Questions?