SISD Architecture (Single Instruction, Single Data)

 SISD Architecture (Single Instruction, Single Data)

Computer Architecture - Unit IV: Parallel Processing

Instituto Tecnológico de Comitán

Instructor: M. en S. Paulo Eduardo Chapela Gómez

Semester: August-December 2025

What is SISD?

  • • SISD stands for Single Instruction, Single Data.
  • • It is the simplest and most traditional computer architecture.
  • • A single processor executes one instruction at a time on one data stream.
  • • Example: A traditional sequential CPU.

How SISD Works

  • • The Control Unit (CU) fetches and decodes one instruction at a time.
  • • The ALU executes that instruction using data from memory.
  • • The result is stored back in memory or a register.
  • • Process: 1. Fetch → 2. Decode → 3. Execute → 4.Store

SISD Diagram

Typical SISD Model:

| Control Unit |

| ALU/CPU Core |

|  Memory |

| Data Stream | 

Examples of SISD Systems
  • • Early computers: IBM 360, Intel 8086, Motorola 68000.
  • • Most modern personal computers (when using a single core).
  • • Sequential programs (e.g., basic C or Java code without threads).
Advantages of SISD
  • 1. Simple design and control.
  • 2. Easier to program - no need for synchronization.
  • 3. Predictable execution time.
  • 4. Lower cost and hardware complexity.
SIMD Architecture
  • Single Instruction, Multiple Data
  • Overview of parallel data processing.
Definition
SIMD refers to a computer architecture in which a single instruction is executed simultaneously on multiple data elements.
Key Characteristics

  • One control unit
  • Multiple processing units
  • Executes the same instruction across multiple data points
  • Highly parallel data processing
Advantages
  • High performance for vector and matrix operations
  • Efficient for graphics, multimedia, and scientific computing
  • Reduces instruction overhead
Disadvantages
  • Not suitable for tasks requiring different operations per data element
  • Less flexible than MIMD architectures
  • Requires data to be highly regular and parallelizable
Common Applications
  • Image and video processing
  • Deep learning and AI acceleration
  • Vectorized numerical computations
  • GPU operations

Comentarios

Publicar un comentario

Entradas más populares de este blog

CHIPSET

Imagen
A chipset is a collection of integrated circuits on a computer's motherboard that manages the flow of data between the processor, memory, and peripheral devices. It functions as a "traffic controller," dictating how components communicate with each other and establishing the capabilities of the motherboard itself. Core functions of a chipset Controls data flow. The chipset acts as the communication hub, ensuring that data moves smoothly and efficiently between components like the Central Processing Unit (CPU), Graphics Processing Unit (GPU), Random Access Memory (RAM), and storage devices. Determines compatibility: The chipset is a key factor in deciding which processors, types of RAM, and expansion cards (e.g., graphics and storage) are compatible with your motherboard. A CPU must be able to communicate with the chipset to function correctly. Manages peripherals: It controls input/output (I/O) devices by managing connections like USB ports, SATA connectors for storage d...
Leer más

Characteristics of RISC

 RISC (Reduced Instruction Set Computer) RISC is the way to make hardware simpler whereas CISC is the single instruction that handles multiple work. In this article, we are going to discuss RISC and CISC in detail as well as the difference between RISC and CISC. Let's proceed with RISC first. Reduced Instruction Set Architecture (RISC) The main idea behind this is to simplify hardware by using an instruction set composed of a few basic steps for loading, evaluating, and storing operations, just like a load command will load data; a store command will store the data. Characteristics of RISC Simpler instruction, hence simple instruction decoding. Instruction comes undersize of one word. Instruction takes a single clock cycle to get executed. More general-purpose registers. Simple Addressing Modes. Fewer Data Types. A pipeline can be achieved.
Leer más