Scribd
Upload
26 views

THE Architecture of Parallel Computers: by Mykhailo Prostak, IA-73

This document discusses the architecture of parallel computers. It covers key hardware issues like the number and type of processors, memory hierarchy, and I/O devices. Operating system issues for parallel computers include allocating resources, multi-processing, I/O management, and accessing hardware features. Programming parallel computers presents difficulties in compiler support, programmability, and designing parallel algorithms. The document also describes Flynn's classification of computer architectures and examples like SISD, SIMD, MIMD, and MISD models.

Uploaded by

U GOU
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
26 views

THE Architecture of Parallel Computers: by Mykhailo Prostak, IA-73

This document discusses the architecture of parallel computers. It covers key hardware issues like the number and type of processors, memory hierarchy, and I/O devices. Operating system issues for parallel computers include allocating resources, multi-processing, I/O management, and accessing hardware features. Programming parallel computers presents difficulties in compiler support, programmability, and designing parallel algorithms. The document also describes Flynn's classification of computer architectures and examples like SISD, SIMD, MIMD, and MISD models.

Uploaded by

U GOU
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13

THE

ARCHITECTURE
OF PARALLEL
COMPUTERS
By Mykhailo Prostak, IA-73
Computer • No Component Software Can be Treated In Isolation
From the Others

Systems
Hardware • Number and Type of Processors
• Processor Control
Issues • Memory Hierarchy
• I/O devices and Peripherals
• Operating System Support
• Applications Software Compatibility
Operating • Allocating and Managing Resources
• Access to Hardware Features
System Issues - Multi-Processing
- Multi-Threading
• I/O Management
• Access to Peripherals
• Efficiency
Applications • Compiler/Linker Support
• Programmability
Issues • OS/Hardware Feature Availability
• Compatibility
• Parallel Compilers
- Preprocessor
- Precompiler
- Parallelizing Compiler
Architecture • Scalar Architecture
• Prefetch Fetch/Execute Overlap
Evolution • Multiple Functional Units
• Pipelining
• Vector Processors
• Lock-Step Processors
• Multi-Processor
Flynn’s • Consider Instruction Streams and Data Streams
Separately

Classification • SISD - Single Instruction, Single Data Stream


• SIMD - Single Instruction, Multiple Data Streams
• MIMD - Multiple Instruction, Multiple Data Streams
• MISD - (rare) Multiple Instruction, Single Data
Stream
SISD • Multiple Processors Execute a Single Program
• Each Processor operates on its own data
• Vector Processors
• Array Processors
• PRAM Theoretical Model
MIMD • Multiple Processors cooperate on a single task
• Each Processor runs a different program
• Each Processor operates on different data
• Many Commercial Examples Exist
MISD • A Single Data Stream passes through multiple
processors
• Different operations are triggered on different
processors
• Systolic Arrays
• Wave-Front Arrays
Programming • Parallel Computers are Difficult to Program
• Automatic Parallelization Techniques are only
Issues Partially Successful
• Programming languages are few, not well
supported, and difficult to use
• Parallel Algorithms are difficult to design
Other • MIPS rate - Millions of instructions per second
• Clock Rate for similar processors
Measures • MFLOPS rate - Millions of floating point operations per
second
• These measures are not neccessarily directly
comparable between different types of processors
THANKS FOR
WATCHING!

You might also like