Computer Architecture Basics
48 Questions
0 Views

Choose a study mode

Play Quiz
Study Flashcards
Spaced Repetition
Chat to lesson

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

The contents of a register can be transferred to the accumulator.

True (A)

Index registers are only used for storage and counting operations.

False (B)

The effective address for indexed addressing is calculated by subtracting the contents of the index register from the address part of the instruction.

False (B)

Base registers are primarily utilized in dynamic relocation and addressing memory spaces exceeding the instruction-type address field capacity.

<p>True (A)</p> Signup and view all the answers

Index registers are automatically incremented or decremented after each use in all cases.

<p>False (B)</p> Signup and view all the answers

Instructions for modifying general-purpose registers are not available to the programmer.

<p>False (B)</p> Signup and view all the answers

The concept of multiple accumulators is only relevant for processors that do not have general-purpose registers.

<p>False (B)</p> Signup and view all the answers

The ALU performs operations on operands stored in registers accessible to programmers.

<p>True (A)</p> Signup and view all the answers

Base registers are a type of general-purpose register that is optimized for memory addressing tasks.

<p>False (B)</p> Signup and view all the answers

The clock signal for the control unit is typically generated within the unit itself.

<p>False (B)</p> Signup and view all the answers

Single operand operations, such as resetting to zero or shifting, involve manipulating two registers simultaneously.

<p>False (B)</p> Signup and view all the answers

The Instruction Decoder decodes the current instruction to determine the appropriate sequence of commands for execution.

<p>True (A)</p> Signup and view all the answers

Floating point and double precision operations were initially a common feature in general-purpose computers, but are now limited to scientific computing.

<p>False (B)</p> Signup and view all the answers

The ALU performs a wide variety of operations, including division and bitwise negation, but does not handle logical operations like AND, OR, and XOR.

<p>False (B)</p> Signup and view all the answers

Accumulator extension is a feature that involves using the results of an ALU operation to further modify the value of a special register.

<p>True (A)</p> Signup and view all the answers

The ALU is responsible for executing instructions within a specific program.

<p>False (B)</p> Signup and view all the answers

Flash Memory, unlike EEPROM, is designed for permanent data storage.

<p>False (B)</p> Signup and view all the answers

Cache memory is a separate component from the main memory and specifically stores important instructions and data required by the CPU.

<p>True (A)</p> Signup and view all the answers

The evolution of ROM has led to types that allow for reprogramming and erasure, making it function similar to RAM.

<p>True (A)</p> Signup and view all the answers

Information in cache memory is accessed through a sequential search based on memory addresses.

<p>False (B)</p> Signup and view all the answers

The concept of "antémémoire" refers to a technology that increases the speed of transferring data between the CPU and main memory.

<p>True (A)</p> Signup and view all the answers

Flash Memory can store data in blocks, leading to slower data transfer rates compared to other storage devices.

<p>False (B)</p> Signup and view all the answers

Cache memory's associative addressing mechanism allows for simultaneous comparison of all memory entries to find the required information.

<p>True (A)</p> Signup and view all the answers

The advancements in memory technologies have little impact on the future of computing, especially for personal electronics and data centers.

<p>False (B)</p> Signup and view all the answers

Near-line storage media libraries typically contain hundreds or thousands of media units.

<p>True (A)</p> Signup and view all the answers

Accessing information in near-line storage requires manual intervention to load the media into a drive.

<p>False (B)</p> Signup and view all the answers

Offline storage media are usually kept in secure storage facilities.

<p>True (A)</p> Signup and view all the answers

The capacity of a memory system is measured in terms of bits or bytes, with bytes being a more common unit.

<p>True (A)</p> Signup and view all the answers

Memory access time refers to the duration required to complete both a read and a write operation.

<p>False (B)</p> Signup and view all the answers

The memory cycle time is shorter than the access time, as it includes additional operations like signal stabilization.

<p>False (B)</p> Signup and view all the answers

Data transfer rate measures the number of information units written per second, but not read per second.

<p>False (B)</p> Signup and view all the answers

Volatile memory retains its contents even when the power is turned off, while magnetic mass storage is volatile.

<p>False (B)</p> Signup and view all the answers

Each character was originally defined using 8 bits.

<p>False (B)</p> Signup and view all the answers

The capacity of memory is expressed by the number of memory cells in central memory.

<p>True (A)</p> Signup and view all the answers

A memory word typically consists of 1 byte.

<p>False (B)</p> Signup and view all the answers

Before 2008, most personal computers could address a maximum of 8 GB of memory.

<p>False (B)</p> Signup and view all the answers

A 64-bit architecture allows for addressing 264 bytes.

<p>True (A)</p> Signup and view all the answers

The log2 value of 1024 is 10 bits.

<p>True (A)</p> Signup and view all the answers

Address registers contain the actual data stored in the memory cells.

<p>False (B)</p> Signup and view all the answers

A 32-bit address register can address approximately 4 billion bytes.

<p>True (A)</p> Signup and view all the answers

The Program Status Word (PSW) does not contain any flags.

<p>False (B)</p> Signup and view all the answers

The Zero Flag (ZF) indicates whether the result of an operation is zero.

<p>True (A)</p> Signup and view all the answers

The Stack Pointer (SP) points to the bottom address of the stack.

<p>False (B)</p> Signup and view all the answers

The stack operates on a Last In First Out (LIFO) principle.

<p>True (A)</p> Signup and view all the answers

The Carry Flag (CF) indicates an overflow condition during an operation.

<p>False (B)</p> Signup and view all the answers

Multiple Stack Pointers (SP) can be used to simulate multiple LIFO stacks for different processes.

<p>True (A)</p> Signup and view all the answers

Information stored in the stack is read in the same order as it was stored.

<p>False (B)</p> Signup and view all the answers

The state register cannot be tested programmatically to determine instruction sequences.

<p>False (B)</p> Signup and view all the answers

Study Notes

Computer's Main Components

  • A computer is composed of central memory, a central processing unit (CPU), and input/output units.
  • Central memory stores programs and data.
  • The CPU executes programs loaded into central memory.
  • Input/output units allow information exchange with peripheral devices.
  • The central unit combines the CPU and central memory.

The Overall Diagram of an Architecture

  • A computer is composed of central memory, a central processing unit (CPU), and input/output units.
  • Central processing unit executes programs loaded into central memory.
  • Input/output units allow the exchange of information with peripheral devices.
  • Central unit is a combination of the CPU and the central memory.

The General Diagram of a Computer

  • Shows the relationship between the CPU, cache memory, main memory, and input/output units.
  • The CPU consists of the control unit and the arithmetic logic unit (ALU).
  • The ALU performs arithmetic and logic operations.
  • The control unit manages the execution of program instructions.
  • Peripheral units connect to input/output units.

Execution of a Program

  • Programs and data load into main memory.
  • Instructions are brought one by one to the control unit.
  • The control unit analyzes instructions and signals the ALU for processing.
  • Processing may involve accessing input/output units or the main memory.
  • A fast cache memory is located between main memory and the CPU.

Central Memory

  • Central memory stores programs and data
  • Data is stored using various encoding schemes, such as ASCII (American Standard Code for Information Interchange).
  • Instructions are stored in machine code.
  • Examples given include machine code for addition on an Intel 8086 microprocessor (10000001) and ASCII encoding for the character "F" (1011001).

Memory Cells

  • Each memory cell has a unique address and contains either an instruction or data.
  • Memory capacity measured in bytes (e.g., 2 GB = 2 × 230≈ 2 billion bytes).
  • Processing operations (read or write) are performed on memory words, typically 4 or 8 bytes.

Data and Memory Registers

  • Modern processors manipulate data more than one byte (32-bit or 64-bit).
  • Memory registers have specific functions.
  • Two types of registers are address registers and word registers.
  • Word registers store memory word contents.
  • Address registers contain memory cell addresses.

Address Registers and Memory Capacity

  • The address register's size depends on the number of memory cells (e.g., log2(1024) = 10 bits for 1024 cells).
  • A 32-bit address register allows addressing 232 different bytes (4GB).
  • 64-bit architectures allow for theoretically addressing 264 bytes ≈ 16 billion billion bytes.

Operations in Central Memory: Reading and Writing

  • Reading: The address register contains the cell address to be read, and the contents are copied into a word register.
  • Writing: The address register contains the cell address where the word register's content will be written (overwriting).

Access Time and RAM

  • The time to write or read a memory word is called access time.
  • Access time varies from a few nanoseconds to a few hundred.
  • If access times are identical for every cell in central memory, the memory is called RAM (Random Access Memory).

Central Processing Unit (CPU)

  • The CPU has two main units: the control unit and the arithmetic and logic unit (ALU).
  • The control unit manages program execution.
  • The ALU performs arithmetic and logic operations.

Control Unit

  • The control unit manages program execution.
  • It has two important registers: the instruction register (IR) and the program counter (PC).
  • The IR holds the current executed instruction.
  • The PC holds the address of the next instruction.

Program Counter and Instruction Execution

  • The program counter (PC) is incremented to get the next instruction address.
  • The increment may vary depending on the instruction size.
  • The control unit's registers aren't directly accessible to programmers.

Control Unit Components

  • Instruction Decoder decodes the current instruction.
  • Command Sequencer activates necessary circuits for instruction execution using clock signals.
  • Clock signals external to the control unit.

Arithmetic and Logic Unit (ALU)

  • The ALU contains circuits for arithmetic operations (addition, subtraction, multiplication, division) and bitwise negation (bit inversion).
  • It also includes logical operations (AND, OR, XOR).
  • Operands are stored in registers within the ALU.
  • These registers are accessible to programmers.

Arithmetic and Logic Unit (ALU) (Continued)

  • ALUs in modern computers can perform various operations.
  • Single operand operations involve one register and one operand (resetting to zero, logical complement, shifting, incrementing).
  • Two operand operations involve two operands (addition, subtraction, logical operations).
  • Accumulator extension can be used in some operations.
  • Floating-point and double-precision operations were historically separate but are now integrated into the ALU.

Processing in the ALU

  • All data processing occurs within the ALU.
  • The ALU contains circuits for fundamental operations in algorithms.
  • The ALU is entirely controlled by the control unit.

Registers of the ALU (Arithmetic Logic Unit)

  • ALU registers are divided into groups: arithmetic registers, base and index registers, and general-purpose registers (for storing intermediate results).

ALU Registers (Continued)

  • The status register (PSW, Program Status Word) indicates the system state (carry, overflow, etc.).

Access Time and Memory Performance

  • Registers have much faster access times than main memory.
  • Computers use mechanisms to partially compensate for slower main memory speeds. This includes more CPU registers and adding faster cache memory between CPU and main memory.
  • Faster cache memory improves the overall speed by reducing accesses to the main, slower memory.

Input/Output Units

  • Input/output (I/O) units transmit information between the CPU and peripheral devices.
  • Typical I/O units include the bus, Direct Memory Access (DMA), and channels.
  • Information exchange is slower as it moves farther from the CPU. Therefore, I/O operations are managed by special processors rather than managed by the CPU directly.

Peripheral Units

  • Peripheral units are broadly classified into devices that exchange data with the outside world, and auxiliary memory devices for permanent storage.
  • Examples of data exchange devices include screens, keyboards, printers.
  • I/O devices have auxiliary memory, that use media like disks, tapes, and cartridges where data is stored permanently.
  • Storage is often slower than main memory but with a higher capacity.

Mass Storage and Controllers

  • Auxiliary memories are also referred to as mass storage.
  • Mass storage enables permanent storage of data (unlike volatile main memory).
  • Each storage device category has an associated device controller. This controller interfaces with the I/O units.

Busses

  • A bus is a set of lines that connects devices.
  • Bus lines transmit addresses, data, or control signals.
  • Busses are shared between devices, but when in use, a device takes control, which in turn means it is reserved to that device.
  • I/O devices use buffer registers and interface circuits for communication with computers and other I/O devices.
  • Buses can be divided into Data, Address and Control busses.

Microcomputer Architecture

  • The data bus allows two-way information transfer.
  • The data bus lines match the bit architecture (number of bits) of the computer. (e.g. 8, 16, 32, 64, 128-bit computer has 8, 16, 32, 64, 128 data bus lines respectively).
  • Number of data bus lines corresponds to the number of pins on the microprocessor dedicated to data.

CPU Registers

  • CPU registers' number and type are important parts of its architecture and heavily influence programming; The type varies between manufacturers.
  • The basic functions performed by registers are generally similar.
  • This information should detail the key registers, their function, and how they are modified by a program.

Instruction Register (IR)

  • Fetches instructions from memory and places them in the IR.
  • Size of the instruction register (IR) corresponds to the size of the instruction.
  • Programmers typically don't access the instruction register directly.
  • Bits in the instruction register are sent to a decoder or to microprogram memory to find out which operation to perform.

Program Counter (PC)

  • Holds the memory address of the next instruction.
  • PC automatically increments after each use, ensuring sequential instruction execution.
  • Instructions like jump or branch will change the PC to a new address, thus altering the sequence.

Program Counter (PC)

  • Program counter (PC) value changes during execution cycle, after decoding the operation code, before transferring it to Memory Address Register (MAR).
  • Size of the PC depends on the addressable memory locations.
  • Programmers don't typically access the PC directly.

Accumulator (ACC)

  • In single-address machines, the accumulator (ACC) is a crucial register in the ALU.
  • ALU operations often use the accumulator.
  • The ACC serves as a buffer in input/output (I/O).
  • Accumulators can have an extension for double-precision operations and more data storage.

Accumulator (ACC)

  • This extension is used for multiplication or division operations (dividends and quotients).
  • Some processors have multiple accumulators. Using instruction codes to specify the use of the correct accumulator.

General Purpose Registers

  • General-purpose registers (scratchpad registers) store frequently used data during program execution.
  • This accelerates execution by reducing memory accesses.
  • Programmers typically use instructions for manipulating them.

General Purpose Registers

  • Typical operations include loading from memory or another register, storing to memory, transferring to the Accumulator (ACC) or from it, and incrementing or decrementing.

Index Registers (XR)

  • Index registers can be used like general-purpose registers but are also useful for manipulating data arrays.
  • Effectively modifying addresses through indexed addressing (modifying an address by adding to the index register).

Index Registers (XR)

  • The effective address of an operand is calculated by adding the instruction's address part to the index register's contents.
  • Instructions are available for incrementing or decrementing index registers, often automatically doing so.

Base Registers

  • Base registers are used similarly to index registers, generating effective addresses via addressing calculated using the contents of the address field (of the instruction) plus a base.
  • Base registers are useful for dynamic relocation.
  • Enables addressing for memory spaces exceeding the capacity of the instruction's address field.

Program Status Word (PSW)/State Register (SR)

  • Also called the condition register, contains flags to indicate CPU status.
  • Flags include Zero Flag (ZF), Carry Flag (CF), and Overflow Flag (VF).
  • Flags used to determine instruction execution flows (e.g., conditional jumps).

Program Status Word (PSW)/State Register (SR)

  • Registers contain bits concerning CPU's state (user/supervisor mode or interrupt status)

Stack Pointer (SP)

  • Simulates a stack in main memory.
  • The stack pointer (SP) is a memory address register (MAR) specifically for the stack portion of RAM.
  • The SP's value is updated as bytes are pushed or popped from the stack.
  • SP points to the stack's top.

Stack Pointer (SP)

  • The Stack Operation (LIFO) or Last-In, First-Out.
  • The stack stores entries in order, and retrieves them in reverse order. The next value stored in the stack is stored at the location after the previous one was stored.

Specialized Registers

  • Some machines have specialized registers (e.g., stack pointers, shift registers, floating-point registers) for specific operations (e.g. bit-manipulation, decimal number operations).
  • Multiple stack pointers can be used when simulating multiple LIFO stacks.

Memory

  • Memory is a device that records, stores, and retrieves information (in binary).
  • Often categorized by capacity and access time into types like CPU registers, cache, main memory, secondary (mass) storage.
  • The hierarchy is illustrated using a pyramid diagram (with speed and capacity axes).

Memory Hierarchy in a Computer

  • Memory components are logically organized by their access time, capacity and cost per bit. Moving further from CPU to mass storage, increases capacity, but decreases the cost per bit and access time gets slower.

High-Speed Registers

  • High-speed registers are located within the CPU.
  • Characterized by extremely fast access times.
  • These are frequently for temporarily holding operands, intermediate results during calculations and thus speeding up processing.

Cache Memory (Antememory)

  • A fast, small-capacity memory that acts as a buffer between the CPU and main memory.
  • Saves time by often reducing the number of instructions to access slower main memory.
  • It's typically much smaller than main memory but significantly faster.

Main Memory (Central Memory)

  • Programs and data loaded into the primary workspace for the CPU (Central Processing Unit).
  • Main memory (in a stored-program computer) is required to hold instructions for execution.
  • Main memory (central memory) is typically semiconductor-based, but has slower access compared to CPU registers or cache.

Mass Storage

  • Primary mass storage includes hard drives and solid-state drives (SSD's) for permanent storage.
  • Slower access times, but drastically higher capacity (permanently stored data).
  • Secondary mass storage is for long-term storage, backups, and archiving and includes devices like magnetic disks, tapes, and optical discs.

Permanent Storage Systems

  • Systems for permanent data storage in primary or secondary mass storage devices.
  • Online storage: fast access using hard or solid state drives and are usually network accessible.
  • Network storage is often categorized by the way they are accessed (locally or through the Internet). Networks attached storage (NAS), storage area network (SAN) and cloud storage are all types for network storage.

Near-Line and Offline Storage

  • Near-line storage uses optical, magneto-optical, or magnetic tape media.
  • Offline storage uses media that is stored on shelves (offline) and needs manual intervention.
  • Usually larger capacities but have slower access times compared to online storage.

Characteristics of Memory

  • Address designates a memory element (e.g., memory location of a byte in main memory).
  • Capacity is the maximum amount of information.
  • Access time is the interval between starting and completing a memory access operation.
  • Memory cycle is the minimum time between successive memory accesses.

Characteristics of Memory (Continued)

  • Data Transfer Rate is the rate at which information is read or written per unit time.

  • Volatility refers to whether memory contents are retained when power is off. Volatile memory loses its contents, while non-volatile memory retains the contents.

Different Types of Memory Access

  • Sequential Access: slowest, must go through all preceding information (e.g., magnetic tapes).
  • Direct Access: information has its own address (e.g., main memory, registers).
  • Semi-Sequential Access: combines direct and sequential access. (e.g., magnetic disks - direct access to cylinder, sequential within cylinder).

Main Memory

  • The definition of main memory function covers both the instructions and data needed by programs to run and the operating system.
  • Instructions are executed sequentially with parts being loaded into main memory from secondary storage and being loaded from there into the CPU for execution.
  • Capacity and access speed are critically important aspects of a CPU's performance.

Technological Evolution of Main Memory

  • Main memory uses technologies like delay lines, vacuum tubes, magnetic drums (primarily in the 1940s and 1950s).
  • Magnetic core memory replaced earlier methods.
  • Semiconductor memory became dominant in the 1970s offering enhanced performance.

Semiconductor Memories

  • Semiconductor memories became the standard around the 1970s and their technology has steadily advanced in terms of capacity, speed, and cost efficiency.
  • Types: SRAM, DRAM, and MRAM.

Read-Write and Read-Only Memory

  • Read-Write Memory (RWM): Allows both reading and writing operations. Also known as "volatile" or "live" memory.
  • Read-Only Memory (ROM): Permits only reading operations. Writing is either not possible, or only under very specific conditions.

Characteristics of Major Types of Semiconductor Memory

  • RAM (Random Access Memory) access time is independent of the memory location.
  • SRAM uses unipolar and bipolar technology (bipolar being slightly faster) for logic that can be much faster while needing more power.
  • DRAM uses MOS (Metal Oxide Semiconductor) and needs refreshing (every few milliseconds), which is slower than SRAM.
  • MRAM uses magnetization and is nonvolatile but still in development.
  • Access time often measured in nanoseconds, and transfer rates measured in GB/s.

Characteristics of Major Types of Semiconductor Memory (cont.)

  • Advantages of DRAM: simpler manufacturing process, high integration density, lower cost per bit.
  • Disadvantages of DRAM: needs refresh logic. Types of DRAM: SDRAM (Synchronous DRAM), and DDR.

VRAM and MRAM

  • VRAM: special type of DRAM used in graphics cards, with two ports for simultaneous read and write.
  • MRAM: uses magnetization to store data and is nonvolatile, though still under development. Nonvolatile memory will retain the data when power is removed.

ROM Memory

  • Read Only Memory (ROM): Non-volatile memory which retains its data even with power off. Data in ROM is typically written during manufacture, therefore normally unchangeable.

ROM Memory (Characteristics)

  • ROM is often expensive to produce due to the complexity of the manufacturing processes such as mask fabrication. Production of ROM involves MOS (Metal-Oxide-Semiconductor) and bipolar technologies. Common applications involve code conversion, character generation and storage for essential system programs.

PROM and EPROM

  • PROM (Programmable ROM): Can be written once by using a suitable program or specialized equipment. Used for applications where no further changes to the stored data are needed.
  • EPROM (Erasable Programmable ROM): Designed for multiple write operations. Erision and reprogramming using ultraviolet light. Used for development work and iterative designs.

EEPROM

  • EEPROM (Electrically Erasable Programmable ROM): Permits electrical erasure and reprogramming of the memory. No need for removal from the computer for this function.
  • Users benefit from faster, more flexible data updates over EPROM.

Flash Memory

  • Flash memory combines the benefits of EEPROM and permanent storage capabilities.
  • It uses floating gate transistors to store data in hundreds of gigabytes.
  • Flash memory allows addressing data in entire blocks rather than individual bytes which leads to increased data transfer rates, and reliability. This makes it well suited for portable devices and widely used in USB drives (Universal Serial Bus) and SSDs (Solid-State Drives).

Evolution of ROM Memory

  • ROM memory has progressed from a purely read-only type of memory to types with the capability of erasure and reprogramming, and similar features akin to RAM.
  • Flash memory, based on EEPROM technology, reflects these advancements. Technological advancements continue to influence the future of computing and impact personal electronic devices.

Cache Memory

  • Cache memory ("antémémoire") addresses the performance gap between CPU and main memory speed by inserting a fast SRAM (Static Random Access Memory).

  • Cache contains frequently used CPU data blocks.

  • Cache memory is associative: data access by matching the content rather than the address. Each block in the cache has a key matching the address from main memory of the data, instructions or data sought by the CPU, and an associated memory location containing the data or instructions.

Searching in Cache Memory

  • Searching in cache memory happens in parallel across all entries.
  • The process determines whether the sought data block is present within the cache.
  • This process uses comparator circuits in each cell in the cache to allow for quick access to the relevant memory location, if the item is present.

Multi-Level Cache

  • Modern computers frequently use multiple levels of cache memory for data and instructions.
  • Each cache is typically a few megabytes or larger.
  • The highest level cache will generally be much faster than the lowest level cache, while the lowest will probably be the largest.
  • Information access and retrieval is driven by the locality principle. The principle describes the high probability of needing to access similar data blocks and instructions in a relatively short period.

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Quiz Team

Related Documents

Description

Test your knowledge on the fundamentals of computer architecture, focusing on registers, addressing modes, and the ALU. This quiz covers key concepts such as index and base registers, their roles, and how they interact within the control unit. Perfect for students learning about computer systems and their components.

More Like This

Use Quizgecko on...
Browser
Open
Browser
Browser