Scribd
Upload
64 views

Coe123 Report

This document discusses multiprocessor systems and fault tolerance. It begins by defining a multiprocessor as a computer with two or more CPUs that cooperate on tasks. Multiprocessors can improve performance, scalability, and reliability. They are classified by memory systems as distributed, shared, massively parallel, or modestly parallel. Shared bus systems provide simplicity and low cost but are slow. Cache coherence is discussed as an issue that occurs when caches contain different versions of the same data. Fault tolerance is the ability of a system to operate correctly despite failures, and it is achieved through redundancy. Approaches include hardware, software, information, time, static, and dynamic redundancy.

Uploaded by

Thesa Fajardo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
64 views

Coe123 Report

This document discusses multiprocessor systems and fault tolerance. It begins by defining a multiprocessor as a computer with two or more CPUs that cooperate on tasks. Multiprocessors can improve performance, scalability, and reliability. They are classified by memory systems as distributed, shared, massively parallel, or modestly parallel. Shared bus systems provide simplicity and low cost but are slow. Cache coherence is discussed as an issue that occurs when caches contain different versions of the same data. Fault tolerance is the ability of a system to operate correctly despite failures, and it is achieved through redundancy. Approaches include hardware, software, information, time, static, and dynamic redundancy.

Uploaded by

Thesa Fajardo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 24

MULTIPROCESSORS

BORJA, KEVIN RICH D.

TOPICS
SHARED BUS SYSTEMS
CACHE COHERENCE
FAULT TOLERANCE

MULTIPROCESSOR
is an MIMD computer containing two or more CPUs
that cooperate on common computational tasks

Main Reasons why Multiprocessors are


created
To improve performance
To improve scalability
Symmetric

To improve reliability

Classification of Multiprocessors
according to memory systems
Distributed Memory (loosely coupled)
Shared-Memory (Tightly coupled)

Classification of Multiprocessors
according to memory systems
Massively Parallel
Modestly Parallel

WHAT IS SHARED-BUS
SYSTEMS?
Reasons why shared-bus systems are created
Simplicity
Low cost
Slow

Uniform Access Memory

Basic Synchronization Problem


CPU1
REGION
R
CPU2

Semaphore (Flag)

CACHE COHERENCE
a problem that occurs when two or more caches of different
CPUs contain different versions of the same information at the
same time.

Ways on how to solve Cache


Coherence
Software
Write-Through policy
Mark (tag) information during program compilation as either cacheable or
noncacheable
Noncacheable
Cacheable

Hardware
Advantage
Higher speed and program transparency
Processor should broadcast its write operations to all caches and the global memory via
shared bus
Cache Snooping

CACHE SNOOPING
Snoop Hit
Course of Actions to take on eliminating Snoop hit
Suppose that CPU1 makes a snoop read hit when its cache copy of
the requested item is dirty and it has not yet updated main
memory (write-back policy).
If CPU1 makes a snoop write hit, it knows that its own cache copy
of the requested item is about to become dirty.

FAULT TOLERANCE
ability of a system to execute specified algorithm
correctly regardless of hardware failures and program
errors [Avizienis 1971]

it is of some concern in all computer systems, while


in applications such as spacecraft control and
telephone switching [Siewiorek and Swarz 1992]

Fault Tolerance = Redundancy


Redundant

Several ways on how


Redundancy is introduced
Hardware Redundancy
Software Redundancy
Information Redundancy
Time Redundancy

Two approaches for designing


fault-tolerant systems
Static redundancy
Dynamic redundancy

Static Redundancy
Refers to the use of redundant hardware or software
components that forms a permanent part of the system
to mask the error signals generated by faults
Voter
n-modular redundancy (nMR)
Example is Triple modular redundancy

Dynamic Redundancy
Three steps in reorganization
Fault Diagnosis
Fault Elimination
Recovery

Redundant Disk Arrays


Drawbacks (Magnetic hard disks)
Relatively Slow data-transfer rates
Electromechanical construction makes them prone to
both transient and catastrophic failures

RAID
Redundant Array of Inexpensive Disks
Main Function
Distribute the stored data over a set of disks configured to
appear like a single large disk

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy