Memory and Storage Systems

Total Page:16

File Type:pdf, Size:1020Kb

Memory and Storage Systems CHAPTER 3 MEMORY AND STORAGE SYSTEMS Chapter Outline Chapter Objectives 3.1 Introduction In this chapter, we will learn: 3.2 Memory Representation ∑ The concept of memory and its 3.3 Random Access Memory representation. 3.3.1 Static RAM ∑ How data is stored in Random Access 3.3.2 Dynamic RAM Memory (RAM) and the various types of 3.4 Read Only Memory RAM. 3.4.1 Programmable ROM ∑ How data is stored in Read Only Memory 3.4.2 Erasable PROM (ROM) and the various types of ROM. 3.4.3 Electrically Erasable PROM ∑ The concept of storage systems and the 3.4.4 Flash ROM various types of storage systems. 3.5 Storage Systems ∑ The criteria for evaluating storage 3.6 Magnetic Storage Systems systems. 3.6.1 Magnetic Tapes 3.6.2 Magnetic Disks 3.7 Optical Storage Systems 3.1 INTRODUCTION 3.7.1 Read only Optical Disks 3.7.2 Write Once, Read Many Disks Computers are used not only for processing of data 3.8 Magneto Optical Systems for immediate use, but also for storing of large 3.8.1 Principle used in Recording Data volume of data for future use. In order to meet 3.8.2 Architecture of Magneto Optical Disks these two specifi c requirements, computers use two 3.9 Solid-State Storage Devices types of storage locations—one, for storing the data 3.9.1 Structure of SSD that are being currently handled by the CPU and the 3.9.2 Advantages of SSD other, for storing the results and the data for future 3.9.3 Disadvantages of SSD use. The storage location where the data are held 3.10 Storage Evaluation Criteria temporarily is referred to as the primary memory Chapter Summary while the storage location where the programs Key Terms to Remember and data are stored permanently for future use is Review Questions referred to as the secondary memory. The primary Fill in the Blanks memory is generally known as “memory” and the secondary memory as “storage”. Multiple Choice Questions The data and instructions stored in the primary Discussion Questions memory can be directly accessed by the CPU 44 Fundamentals of Computers using the data and address buses. However, the information stored in the secondary memory is not directly accessible to CPU. Firstly, the information has to be transferred to the primary memory using I/O channels and then, to the CPU. Computers also use a third type of storage location known as the internal process memory. This memory is placed either inside the CPU or near the CPU (connected through special fast bus). Figure 3.1 illustrates all the three categories of computer memory and their relative speed, storage capacity and cost. Fig. 3.1 Memory categories Primary memory (also known as main memory) includes two types, namely, Random Access Memory (RAM) and Read Only Memory (ROM). The data stored in RAM are lost when the power is switched off and therefore, it is known as volatile memory. However, the data stored in ROM stay permanently even after the power is switched off and therefore ROM is a non-volatile memory. Secondary memory (also known as auxiliary memory) includes primarily magnetic disks and magnetic tapes. These storage devices have much larger storage capacity than the primary memory. Information stored on such devices remains permanent (until we remove it). Internal process memory usually includes cache memory and registers both of which store data temporarily and are accessible directly by the CPU. This memory is placed inside or near the CPU for the fast access of data. We have discussed briefl y cache memory and various registers used in Chapter 2. In this chapter, we shall discuss the various types of memory units, RAM and ROM and their functions. We shall also discuss the classifi cation of storage devices such as magnetic disks, magnetic tapes and optical disks and their functions. 3.2 MEMORY REPRESENTATION As we discussed earlier, data being worked on is stored in the computer memory. In the memory, values are represented by sequences of binary digits, known as bits. Most computers use a group of eight bits, Memory and Storage Systems 45 known as a byte, to represent a character. How does the computer know what any particular sequence of bits represents? We can think of memory as a “bunch” of bytes or cells into which we can place data. Each cell, known as a data item, is assigned a unique number known as address (which is like the index in an array). The CPU can identify each cell by its address as shown in Fig. 3.2. Fig. 3.2 Memory representation The byte is defi ned as the “smallest addressable unit” of memory. Most computers use groups of bytes, usually 2 or 4, known as “words” to represent information. Computer memories are often rated in terms of their capacity to store information. Typically, capacities are described using the unit of byte as follows: 1 KB (Kilobyte) = 1,024 bytes 1 MB (Megabyte) = 1,048,576 bytes 1 GB (Gigabyte) = 1,073,741,824 bytes 1 TB (Terabyte) = 1,099,511,627,776 bytes 46 Fundamentals of Computers 3.3 RANDOM ACCESS MEMORY Random Access Memory (RAM) is a volatile memory and loses all its data when the power is switched off. It is the main memory of the computer system that stores the data temporarily and allows the data to be accessed in any order. As compared to the secondary storage, the data can be accessed at a faster speed in RAM because it is the internal memory of the computer. Figure 3.3 shows RAM with ICs. Fig. 3.3 Random access memory with ICs RAM is made up of different ICs, which are mounted on a printed circuit board. RAM stores the application programs and the data on which the user is currently working so that the processor can easily access the required application program and data in a less amount of time. RAM is also known as read/write memory because it can perform both read as well as write operations. The speed of RAM is faster than the other memory devices, such as hard disk, fl oppy disk, etc. The programs, which are being currently executed by the computer system, are stored in RAM. RAM is volatile and, therefore, the programs and the data stored in the RAM get lost when the power supply is switched off. The storage capacity of RAM is usually less than the secondary storage devices. RAM can be categorised into two main types, namely, Static RAM (SRAM) and Dynamic RAM (DRAM), which can be further divided into various types for storing data. Figure 3.4 shows the hierarchy of RAM memory. Fig. 3.4 Types of random access memory Memory and Storage Systems 47 3.3.1 Static RAM Static RAM (SRAM) is a type of RAM in which data is stored till the power of the computer system is switched on. SRAM uses a number of transistors to store a single bit of digital information. Figure 3.5 shows the organisation of data in a cell of SRAM. In the fi gure, b1 and b2 represent the two bit lines and Wx represents the word line. In the memory, the data is stored in the form of a two- dimensional array, containing rows Fig. 3.5 Organisation of data in a cell of static random access memory and columns. The row can be accessed globally by a line (known as word line) and the column can be accessed individually by a line (known as bit line). Depending on the function performed by SRAM, it can be divided into the following three types: ∑ Asynchronous SRAM (ASRAM) ASRAM performs its operations without the use of system clock. It makes use of three signals for working, namely, Chip Select (CS), Write Enable (WE) and Output Enable (OE). The CS signal enables the processor to select the memory for performing read and write operations. If the value of CS signal equals zero, then the memory is enabled to perform the operations. On the other hand, if the value of the CS signal equals one, then the memory is disabled and operations—such as reading and writing in ASRAM—cannot be performed. The signal WE makes the decisions related to data, i.e., whether it should be read from or write to the memory. If the value of WE signal equals zero, then no data can be read from or written to the memory. The signal OE is an active low signal that enables the processor to give the output for the data. If the value of OE signal equals zero, then only it will output the data. ∑ Burst SRAM (BSRAM) BSRAM works in association with the system clock and is also known as synchronous SRAM. BSRAM is most commonly used with high-speed applications because the read and write cycles are synchronised with the clock cycles of the processor. The access-waiting time gets reduced after the read and write cycles are synchronised with the clock cycles. The speed and the cost of BSRAM increases or decreases simultaneously. ∑ Pipeline Burst SRAM (PBSRAM) PBSRAM makes use of the pipeline technology in which a large amount of data is broken up in the form of different packets containing data. These packets are arranged in a sequential manner in the pipeline and are sent to the memory simultaneously. PBSRAM can handle a large amount of data at a very high speed. It is the fastest type of SRAM because it can operate at bus rates as high as 66 MHz.
Recommended publications
  • Fast Non-Volatile RAM Products Superior Price and Performance from a Source You Can Trust
    Fast Non-Volatile RAM Products Superior price and performance from a source you can trust www.freescale.com/MRAM The Future of Non-Volatile Read/Write Memory is Magnetic Freescale Semiconductor delivers the world’s non-volatile for greater than 20 years. Commercial MRAM Products Selector Guide Freescale MRAM Features first commercial magnetoresistive random SRAM-compatible packaging assures Part Number Density Configuration Voltage Speed Grades Temperature Package RoHS Compliant • 35 ns read/write cycle time access memory (MRAM) products. Our alternate sourcing from other suppliers. MR2A16ATS35C 4 Mb 256 Kb x 16 3.3V 35 ns 0 to 70°C 44-TSOP Type II √ • Unlimited read/write endurance MRAM products store data using magnetic MR1A16AYS35 2 Mb 128 Kb x 16 3.3V 35 ns 0 to 70°C 44-TSOP Type II √ • 3.3V ± 10 percent power supply polarization rather than electric charge. Extended Temperature Range and MR0A16AYS35 1 Mb 64 Kb x 16 3.3V 35 ns 0 to 70°C 44-TSOP Type II √ • Always non-volatile with greater than MRAM stores data for decades while reading Superior Reliability 20-year retention and writing at SRAM speed without wear- MRAM delivers a 3 volt high-density out. MRAM products use small, simple • Magnetically shielded to greater than non-volatile RAM that operates over extended cells to deliver the highest density and best 25 oersted (Oe) temperature. MRAM does not exhibit the price/performance in the non-volatile RAM • Commercial, Industrial and Extended charge storage failure modes that limit the marketplace. With our new expanded product Temperature Options data retention or endurance of line, we serve a growing portion of the other technologies.
    [Show full text]
  • Magnetic Storage- Magnetic-Core Memory, Magnetic Tape,RAM
    Magnetic storage- From magnetic tape to HDD Juhász Levente 2016.02.24 Table of contents 1. Introduction 2. Magnetic tape 3. Magnetic-core memory 4. Bubble memory 5. Hard disk drive 6. Applications, future prospects 7. References 1. Magnetic storage - introduction Magnetic storage: Recording & storage of data on a magnetised medium A form of „non-volatile” memory Data accessed using read/write heads Widely used for computer data storage, audio and video applications, magnetic stripe cards etc. 1. Magnetic storage - introduction 2. Magnetic tape 1928 Germany: Magnetic tape for audio recording by Fritz Pfleumer • Fe2O3 coating on paper stripes, further developed by AEG & BASF 1951: UNIVAC- first use of magnetic tape for data storage • 12,7 mm Ni-plated brass-phosphorus alloy tape • 128 characters /inch data density • 7000 ch. /s writing speed 2. Magnetic tape 2. Magnetic tape 1950s: IBM : patented magnetic tape technology • 12,7 mm wide magnetic tape on a 26,7 cm reel • 370-730 m long tapes 1980: 1100 m PET –based tape • 18 cm reel for developers • 7, 9 stripe tapes (8 bit + parity) • Capacity up to 140 MB DEC –tapes for personal use 2. Magnetic tape 2014: Sony & IBM recorded 148 Gbit /squareinch tape capacity 185 TB! 2. Magnetic tape Remanent structural change in a magnetic medium Analog or digital recording (binary storage) Longitudinal or perpendicular recording Ni-Fe –alloy core in tape head 2. Magnetic tape Hysteresis in magnetic recording 40-150 kHz bias signal applied to the tape to remove its „magnetic history” and „stir” the magnetization Each recorded signal will encounter the same magnetic condition Current in tape head proportional to the signal to be recorded 2.
    [Show full text]
  • Archiving Online Data to Optical Disk
    ARCHIVING ONLINE DATA TO OPTICAL DISK By J. L. Porter, J. L. Kiesler, and D. A. Stedfast U.S. GEOLOGICAL SURVEY Open-File Report 90-575 Reston, Virginia 1990 U.S. DEPARTMENT OF THE INTERIOR MANUEL LUJAN, JR., Secretary U.S. GEOLOGICAL SURVEY Dallas L. Peck, Director For additional information Copies of this report can be write to: purchased from: Chief, Distributed Information System U.S. Geological Survey U.S. Geological Survey Books and Open-File Reports Section Mail Stop 445 Federal Center, Bldg. 810 12201 Sunrise Valley Drive Box 25425 Reston, Virginia 22092 Denver, Colorado 80225 CONTENTS Page Abstract ............................................................. 1 Introduction ......................................................... 2 Types of optical storage ............................................... 2 Storage media costs and alternative media used for data archival. ......... 3 Comparisons of storage media ......................................... 3 Magnetic compared to optical media ............................... 3 Compact disk read-only memory compared to write-once/read many media ................................... 6 Erasable compared to write-once/read many media ................. 7 Paper and microfiche compared to optical media .................... 8 Advantages of write-once/read-many optical storage ..................... 8 Archival procedure and results ........................................ 9 Summary ........................................................... 13 References ..........................................................
    [Show full text]
  • The Future of Data Storage Technologies
    International Technology Research Institute World Technology (WTEC) Division WTEC Panel Report on The Future of Data Storage Technologies Sadik C. Esener (Panel Co-Chair) Mark H. Kryder (Panel Co-Chair) William D. Doyle Marvin Keshner Masud Mansuripur David A. Thompson June 1999 International Technology Research Institute R.D. Shelton, Director Geoffrey M. Holdridge, WTEC Division Director and ITRI Series Editor 4501 North Charles Street Baltimore, Maryland 21210-2699 WTEC Panel on the Future of Data Storage Technologies Sponsored by the National Science Foundation, Defense Advanced Research Projects Agency and National Institute of Standards and Technology of the United States government. Dr. Sadik C. Esener (Co-Chair) Dr. Marvin Keshner Dr. David A. Thompson Prof. of Electrical and Computer Director, Information Storage IBM Fellow Engineering & Material Sciences Laboratory Research Division Dept. of Electrical & Computer Hewlett-Packard Laboratories International Business Machines Engineering 1501 Page Mill Road Corporation University of California, San Diego Palo Alto, CA 94304-1126 Almaden Research Center 9500 Gilman Drive Mail Stop K01/802 La Jolla, CA 92093-0407 Dr. Masud Mansuripur 650 Harry Road Optical Science Center San Jose, CA 95120-6099 Dr. Mark H. Kryder (Co-Chair) University of Arizona Director, Data Storage Systems Center Tucson, AZ 85721 Carnegie Mellon University Roberts Engineering Hall, Rm. 348 Pittsburgh, PA 15213-3890 Dr. William D. Doyle Director, MINT Center University of Alabama Box 870209 Tuscaloosa, AL 35487-0209 INTERNATIONAL TECHNOLOGY RESEARCH INSTITUTE World Technology (WTEC) Division WTEC at Loyola College (previously known as the Japanese Technology Evaluation Center, JTEC) provides assessments of foreign research and development in selected technologies under a cooperative agreement with the National Science Foundation (NSF).
    [Show full text]
  • 3 Secondary Storage.PDF
    CAPE COMPUTER SCIENCE SECONDARY STORAGE Secondary storage is needed 1. because there is a limit on the size of primary memory (due to cost) and 2. because RAM is volatile and so data needed for future use must be stored somewhere else so that it can be retrieved when necessary. Secondary storage is also used for backup and archives. When we consider secondary storage devices we must bear in mind the following characteristics of each device : Capacity Access speed Access method and portability Floppy Disk This is a 3.5 inch magnetic disk of flexible material which until recently was a standard feature on most microcomputers. Typically it stores 1.44 MB of data. It’s a thin plastic circle coated with a magnetic material and encased in a rigid plastic to protect it. A metal sliding access shuttle opens when the disk is in the machine allowing the read/write head access to the disk. Data can be written to and erased from a floppy disk. A write protect tab can be used to prevent accidental overwriting of data. Before data can be written to a disk, it must be formatted. This prepares the disk for use by creating a magnetic map on the disks surface. This map consists of tracks and sectors. Formatting also prepares the file allocation table (FAT). The address of a file on a floppy disk is comprised of the track number and the sector number. Floppy disks are direct access devices but they are slow compared to hard disks. The floppies great advantage has been its use as a device to help transport small files between machines.
    [Show full text]
  • Computer Files & Data Storage
    STORAGE & FILE CONCEPTS, UTILITIES (Pages 6, 150-158 - Discovering Computers & Microsoft Office 2010) I. Computer files – data, information or instructions residing on secondary storage are stored in the form of a file. A. Software files are also called program files. Program files (instructions) are created by a computer programmer and generally cannot be modified by a user. It’s important that we not move or delete program files because your computer requires them to perform operations. Program files are also referred to as “executables”. 1. You can identify a program file by its extension:“.EXE”, “.COM”, “.BAT”, “.DLL”, “.SYS”, or “.INI” (there are others) or a distinct program icon. B. Data files - when you select a “save” option while using an application program, you are in essence creating a data file. Users create data files. 1. File naming conventions refer to the guidelines followed while assigning file names and will vary with the operating system and application in use (see figure 4-1). File names in Windows 7 may be up to 255 characters, you're not allowed to use reserved characters or certain reserved words. File extensions are used to identify the application that was used to create the file and format data in a manner recognized by the source application used to create it. FALL 2012 1 II. Selecting secondary storage media A. There are three type of technologies for storage devices: magnetic, optical, & solid state, there are advantages & disadvantages between them. When selecting a secondary storage device, certain factors should be considered: 1. Capacity - the capacity of computer storage is expressed in bytes.
    [Show full text]
  • Unit 5: Memory Organizations
    Memory Organizations Unit 5: Memory Organizations Introduction This unit considers the organization of a computer's memory system. The characteristics of the most important storage technologies are described in detail. Basically memories are classified as main memory and secondary memory. Main memory with many different categories are described in Lesson 1. Lesson 2 focuses the secondary memory including the details of floppy disks and hard disks. Lesson 1: Main Memory 1.1 Learning Objectives On completion of this lesson you will be able to : • describe the memory organization • distinguish between ROM, RAM, PROM, EEPROM and • other primary memory elements. 1.2 Organization Computer systems combine binary digits to form groups called words. The size of the word varies from system to system. Table 5.1 illustrates the current word sizes most commonly used with the various computer systems. Two decades ago, IBM introduced their 8-bit PC. This was Memory Organization followed a few years later by the 16-bit PC AT microcomputer, and already it has been replaced with 32- and 64-bit systems. The machine with increased word size is generally faster because it can process more bits of information in the same time span. The current trend is in the direction of the larger word size. Microcomputer main memories are generally made up of many individual chips and perform different functions. The ROM, RAM, Several types of semi- PROM, and EEPROM memories are used in connection with the conductor memories. primary memory of a microcomputers. The main memory generally store computer words as multiple of bytes; each byte consisting of eight bits.
    [Show full text]
  • SŁOWNIK POLSKO-ANGIELSKI ELEKTRONIKI I INFORMATYKI V.03.2010 (C) 2010 Jerzy Kazojć - Wszelkie Prawa Zastrzeżone Słownik Zawiera 18351 Słówek
    OTWARTY SŁOWNIK POLSKO-ANGIELSKI ELEKTRONIKI I INFORMATYKI V.03.2010 (c) 2010 Jerzy Kazojć - wszelkie prawa zastrzeżone Słownik zawiera 18351 słówek. Niniejszy słownik objęty jest licencją Creative Commons Uznanie autorstwa - na tych samych warunkach 3.0 Polska. Aby zobaczyć kopię niniejszej licencji przejdź na stronę http://creativecommons.org/licenses/by-sa/3.0/pl/ lub napisz do Creative Commons, 171 Second Street, Suite 300, San Francisco, California 94105, USA. Licencja UTWÓR (ZDEFINIOWANY PONIŻEJ) PODLEGA NINIEJSZEJ LICENCJI PUBLICZNEJ CREATIVE COMMONS ("CCPL" LUB "LICENCJA"). UTWÓR PODLEGA OCHRONIE PRAWA AUTORSKIEGO LUB INNYCH STOSOWNYCH PRZEPISÓW PRAWA. KORZYSTANIE Z UTWORU W SPOSÓB INNY NIŻ DOZWOLONY NA PODSTAWIE NINIEJSZEJ LICENCJI LUB PRZEPISÓW PRAWA JEST ZABRONIONE. WYKONANIE JAKIEGOKOLWIEK UPRAWNIENIA DO UTWORU OKREŚLONEGO W NINIEJSZEJ LICENCJI OZNACZA PRZYJĘCIE I ZGODĘ NA ZWIĄZANIE POSTANOWIENIAMI NINIEJSZEJ LICENCJI. 1. Definicje a."Utwór zależny" oznacza opracowanie Utworu lub Utworu i innych istniejących wcześniej utworów lub przedmiotów praw pokrewnych, z wyłączeniem materiałów stanowiących Zbiór. Dla uniknięcia wątpliwości, jeżeli Utwór jest utworem muzycznym, artystycznym wykonaniem lub fonogramem, synchronizacja Utworu w czasie z obrazem ruchomym ("synchronizacja") stanowi Utwór Zależny w rozumieniu niniejszej Licencji. b."Zbiór" oznacza zbiór, antologię, wybór lub bazę danych spełniającą cechy utworu, nawet jeżeli zawierają nie chronione materiały, o ile przyjęty w nich dobór, układ lub zestawienie ma twórczy charakter.
    [Show full text]
  • Qt2t0089x7 Nosplash Af5b9567
    UNIVERSITY OF CALIFORNIA Los Angeles Interface Engineering of Voltage-Controlled Embedded Magnetic Random Access Memory A dissertation submitted in partial satisfaction of the requirements for the degree Doctor of Philosophy in Electrical and Computer Engineering by Xiang Li 2018 © Copyright by Xiang Li 2018 ABSTRACT OF THE DISSERTATION Interface Engineering of Voltage-Controlled Embedded Magnetic Random Access Memory by Xiang Li Doctor of Philosophy in Electrical and Computer Engineering University of California, Los Angeles, 2018 Professor Kang Lung Wang, Chair Magnetic memory that utilizes spin to store information has become one of the most promising candidates for next-generation non-volatile memory. Electric-field-assisted writing of magnetic tunnel junctions (MTJs) that exploits the voltage-controlled magnetic anisotropy (VCMA) effect offers great potential for high density and low power memory applications. This emerging Magnetoelectric Random Access Memory (MeRAM) based on the VCMA effect has been investigated due to its lower switching current, compared with traditional current-controlled devices utilizing spin transfer torque (STT) or spin-orbit torque (SOT) for magnetization switching. It is of great promise to integrate MeRAM into the advanced CMOS back-end-of-line (BEOL) processes for on-chip embedded applications, and enable non-volatile electronic systems with low static power dissipation and instant-on operation capability. To achieve the full potential of MeRAM, it is critical to design magnetic materials with high voltage-induced ii writing efficiency, i.e. VCMA coefficient, to allow for low write energy, low write error rate, and high density MeRAM at advanced nodes. In this dissertation, we will first discuss the advantage of MeRAM over other memory technologies with a focus on array-level memory performance, system-level 3D integration, and scaling at advanced nodes.
    [Show full text]
  • The Effects of Magnetic Fields on Magnetic Storage Media Used in Computers
    NITED STATES ARTMENT OF MMERCE NBS TECHNICAL NOTE 735 BLICATION of Magnetic Fields on Magnetic Storage Media Used in Computers — NATIONAL BUREAU OF STANDARDS The National Bureau of Standards^ was established by an act of Congress March 3, 1901. The Bureau's overall goal is to strengthen and advance the Nation's science and technology and facilitate their effective application for public benefit. To this end, the Bureau conducts research and provides: (1) a basis for the Nation's physical measure- ment system, (2) scientific and technological services for industry and government, (3) a technical basis for equity in trade, and (4) technical services to promote public safety. The Bureau consists of the Institute for Basic Standards, the Institute for Materials Research, the Institute for Applied Technology, the Center for Computer Sciences and Technology, and the Office for Information Programs. THE INSTITUTE FOR BASIC STANDARDS provides the central basis within the United States of a complete and consistent system of physical measurement; coordinates that system with measurement systems of other nations; and furnishes essential services leading to accurate and uniform physical measurements throughout the Nation's scien- tific community, industry, and commerce. The Institute consists of a Center for Radia- tion Research, an Office of Measurement Services and the following divisions: Applied Mathematics—Electricity—Heat—Mechanics—Optical Physics—Linac Radiation^—Nuclear Radiation^—Applied Radiation-—Quantum Electronics' Electromagnetics^—Time and Frequency'—Laboratory Astrophysics'—Cryo- genics'. THE INSTITUTE FOR MATERIALS RESEARCH conducts materials research lead- ing to improved methods of measurement, standards, and data on the properties of well-characterized materials needed by industry, commerce, educational institutions, and Government; provides advisory and research services to other Government agencies; and develops, produces, and distributes standard reference materials.
    [Show full text]
  • EE 4504 Computer Organization Overview
    Overview Historically, the limiting factor in a computer’s performance has been memory access time – Memory speed has been slow compared to the speed of the processor EE 4504 – A process could be bottlenecked by the memory Computer Organization system’s inability to “keep up” with the processor Our goal in this section is to study the development of an effective memory Section 3 organization that supports the processing Computer Memory power of the CPU – General memory organization and performance – “Internal” memory components and their use – “External” memory components and their use Reading: Text, chapters 4 and 5 EE 4504 Section 3 1 EE 4504 Section 3 2 1 Terminology Capacity: the amount of information that Access time: can be contained in a memory unit -- – For RAM, the time to address the unit and usually in terms of words or bytes perform the transfer Word: the natural unit of organization in – For non-random access memory, the time to position the R/W head over the desired location the memory, typically the number of bits used to represent a number Memory cycle time: Access time plus any other time required before a second access Addressable unit: the fundamental data can be started element size that can be addressed in the memory -- typically either the word size or Access technique: how are memory individual bytes contents accessed – Random access: Unit of transfer: The number of data » Each location has a unique physical address elements transferred at a time -- usually » Locations can be accessed in any order and bits in main
    [Show full text]
  • Dissertation
    ADAM: A Decentralized Parallel Computer Architecture Featuring Fast Thread and Data Migration and a Uniform Hardware Abstraction by Andrew “bunnie” Huang Submitted to the Department of Electrical Engineering and Computer Science in partial fulfillment of the requirements for the degree of Doctor of Philosophy at the MASSACHUSETTS INSTITUTE OF TECHNOLOGY June 2002 ­c Massachusetts Institute of Technology 2002. All rights reserved. Author........................................................................... Department of Electrical Engineering and Computer Science May 24, 2002 Certifiedby....................................................................... Thomas F. Knight, Jr. Senior Research Scientist Thesis Supervisor Accepted by ...................................................................... Arthur C. Smith Chairman, Department Committee on Graduate Students 2 ADAM: A Decentralized Parallel Computer Architecture Featuring Fast Thread and Data Migration and a Uniform Hardware Abstraction by Andrew “bunnie” Huang Submitted to the Department of Electrical Engineering and Computer Science on May 24, 2002, in partial fulfillment of the requirements for the degree of Doctor of Philosophy Abstract The furious pace of Moore’s Law is driving computer architecture into a realm where the the speed of light is the dominant factor in system latencies. The number of clock cycles to span a chip are increasing, while the number of bits that can be accessed within a clock cycle is decreasing. Hence, it is becoming more difficult to hide latency. One alternative solution is to reduce latency by migrating threads and data, but the overhead of existing implementations has previously made migration an unserviceable solution so far. I present an architecture, implementation, and mechanisms that reduces the overhead of mi- gration to the point where migration is a viable supplement to other latency hiding mechanisms, such as multithreading.
    [Show full text]