Chapter 1 Compu__Architecture.pdf

Computer Architecture Mouna Bouzazi Polygon University September 30, 2024 Professor September 30, 2024 1 / 20 Plan 1. Introduction 2. Computer Architecture 3. Von Neumann and Harvard Architectures 4. Vo...

Computer Architecture Mouna Bouzazi Polygon University September 30, 2024 Professor September 30, 2024 1 / 20 Plan 1. Introduction 2. Computer Architecture 3. Von Neumann and Harvard Architectures 4. Von Neumann Machine 5. Harvard Architecture 6. Conclusion Professor September 30, 2024 2 / 20  Introduction Introduction to Computers To understand how a computer works, you must first know its definition, as well as its architecture and the evolution of this machine. We will also see in this first chapter, the two most known computer architectures used in computers: Von Neumann and Harvard. Computer science, a contraction of information and automatic, designates all the sciences and techniques related to the automatic processing of information. This processing is carried out by a system, whether concrete (machine) or abstract. Professor September 30, 2024 3 / 20  Introduction What is a computer? The computer is a programmable electronic machine used for processing information encoded in digital form. It can receive input data → ”input function”. Store or perform operations on this data according to a program → ”processing function”. And finally provide results as output → ”output function”. Professor September 30, 2024 4 / 20  Introduction Information Processed by a Computer The information processed by the computer can appear in the form of: numbers, text, sound, drawing or graphics, image... but also instructions that make up a program. This information is represented (encoded) in the form of sequences of binary digits 0 and 1. Professor September 30, 2024 5 / 20  Computer Architecture What is Computer Architecture? Computer architecture is the discipline that corresponds to how the components of a computer system are designed. In computer science, the term architecture refers to the organization of the elements in a system and the relationships between these elements. There are two types of architecture: Hardware Architecture: Deals with the organization of the different physical devices found in a computer. → Logical functioning of each component and communication between components. Software Architecture: Deals with the organization of different programs with each other. → Information encoding and the machine’s instruction set, i.e., the set of operations the machine can execute. Professor September 30, 2024 6 / 20  Computer Architecture Generations of Computers (1) Generation Characteristics 1st Generation (1945-55) Mechanical computers - Technology based on lamps, relays, vacuum tubes, resistors. First electronic calculators. 1946: EDVAC (Electronic Discrete Variable Automatic Compute) by Von Neumann, operational in 1951 (program and data stored in memory). 2nd Generation (1955-65) Transistor-based computers - Transistor technology (replacing vacuum tubes). Emergence of advanced programming languages. 1955: IBM 650, the first mass- produced computer. 3rd Generation (1965-71) Integrated Circuit Computers - Technology of Integrated Circuits (chips) using SSI/MSI (Small Scale Integration/Medium Scale Integration) allowing the placement of a large number of transistors on a single silicon chip. Rise of complex operating systems. 1971: Kenback 1, the first microcomputer. Professor September 30, 2024 7 / 20  Computer Architecture Generations of Computers (2) Generation Characteristics 4th Generation (1971-77) Microcomputing LSI (Large Scale Integration) technology. Development of machine networks. 1976: Apple I (S. Wozniak & S. Jobs) with a keyboard. 5th Generation (1977 onwards) Ubiquitous computers - VLSI/ULSI (Very Large/Ultra Large Scale Integration) technologies, integrating thousands to billions of transistors on a single chip. 1981: PC (Personal Computer) by IBM. Since 1990 New tools - Miniaturization of hardware components, with the advent of nanotechnology. 2007: iPhone, the first smartphone by Apple. Professor September 30, 2024 8 / 20  Von Neumann and Harvard Architectures Von Neumann and Harvard Architectures There are two computer architectures that differ in how they access memory: the Von Neumann Architecture and the Harvard Architecture. In the Von Neumann architecture, programs and data are stored in the same memory and managed by the same information processing subsystem. In contrast, in the Harvard architecture, programs and data are stored and managed by different subsystems. Professor September 30, 2024 9 / 20  Von Neumann Machine Von Neumann Machine At the end of 1946, John Von Neumann, a physicist and mathematician of Hungarian origin, proposed a model of a computer that abstracted from the program and initiated the construction of an EDVAC (Electronic Discrete Variable Automatic Computer). He introduced two new concepts in the digital processing of information: The stored-program concept: Programs and data are stored in the same memory. The use of binary digits (bits) for processing information. Professor September 30, 2024 10 / 20  Von Neumann Machine Von Neumann Machine: Key Concepts a. Stored Program Von Neumann had the idea of using the computer’s memory to store programs, hence the name ”stored-program machine” given to this new type of computer. b. Conditional Branching Von Neumann proposed automating logical decision operations by equipping the machine with an instruction called conditional branching (or conditional sequence break). Professor September 30, 2024 11 / 20  Von Neumann Machine Description of the Von Neumann Architecture Von Neumann defined the architecture of modern computers in 1944, which is still widely used today (only the technologies have changed). In this architecture, a single memory is used for both programs and data, while the central processing unit (CPU) consists of a control unit and an arithmetic and logic unit. The architecture includes a data bus (for programs and data) and an address bus (for programs and data). Professor September 30, 2024 12 / 20  Von Neumann Machine Von Neumann Architecture The Von Neumann architecture decomposes the computer into four parts: Arithmetic and Logic Unit (ALU): Performs calculations (basic operations). Control Unit (CU): Directs other units. It is responsible for: Sequencing operations Sending control signals to other units Supervising the operation of the ALU Sending clock signals to other units Memory: Storage device for information (data and programs). Input-Output Devices: Allow the exchange of information with external devices. Professor September 30, 2024 13 / 20  Von Neumann Machine Components of the Von Neumann Architecture The various components of the system are connected by communication pathways called buses (address bus and data bus). This architecture is used to construct all computers, from nanoprocessors (found, for example, in washing machines) to supercomputers (used for intensive calculations). Professor September 30, 2024 14 / 20  Harvard Architecture Harvard Machine The name of this structure comes from Harvard University, where such an architecture was first implemented with the Harvard Mark I, created by Howard Aiken and built by IBM in 1944. It was also referred to by IBM as the Automatic Sequence Controlled Calculator (ASCC). It was the first computer to use separate memory systems for data and instructions. Professor September 30, 2024 15 / 20  Harvard Architecture Description of the Harvard Architecture In the so-called Harvard architecture (developed at the American university in the 1930s), the program memory is systematically separated from the data memory: the addressing of these memories is independent. A simple Harvard architecture consists of a data bus, a program bus, and two address buses. Data exchanges occur simultaneously between the central processing unit and the two memories, allowing for great flexibility in the storage and use of data. Moreover, the program memory is also partially used as data memory to provide even more processing capabilities with complex algorithms. Professor September 30, 2024 16 / 20  Harvard Architecture Harvard Architecture Usage The Harvard architecture is typically used only in specialized systems or for very specific applications. It is employed in specialized digital signal processing (DSP), typically for video and audio data processing. Additionally, it is used in many small microcontrollers in various electronic applications. Professor September 30, 2024 17 / 20  Harvard Architecture Difference Between Von Neumann and Harvard Architecture (Part 1) Difference Von Neumann Architecture Harvard Architecture Name Named after mathematician and Named after the ”Harvard Mark computer scientist John Von I,” an early relay computer Neumann developed at Harvard University Design The design of the Von Neumann The design of the Harvard architecture is simple. architecture is complex. Memory System Requires a single memory for Requires two separate memories instructions and data. for instructions and data. Bus System Requires only one bus for Requires separate buses for instructions and data. instructions and data. Table 1 – Comparison of Von Neumann and Harvard Architectures (Part 1). Professor September 30, 2024 18 / 20  Harvard Architecture Difference Between Von Neumann and Harvard Architecture (Part 2) Difference Von Neumann Architecture Harvard Architecture Instruction Processing The processor needs two clock The processor can complete an cycles to complete an instruction. instruction in one cycle. Performance Lower performance compared to Easier to pipeline, thus higher Harvard architecture. performance can be achieved. Cost Lower cost. Relatively higher cost. Usage Primarily used in all machines Concept primarily used in (desktop computers, laptops, and microcontrollers and digital high-performance workstations). signal processing (DSP). Table 2 – Comparison of Von Neumann and Harvard Architectures (Part 2). Professor September 30, 2024 19 / 20  Conclusion Concclusion In this introductory chapter, we have explained the definition of a computer and the meaning of the term ’computer architecture’. We have explored how the architecture of modern computers has evolved, and we have discussed the von Neumann and Harvard architectures. In the next chapter, we will delve into the main components of a computer in detail. Professor September 30, 2024 20 / 20