Chapters 8 - PowerPoint PPT Presentation

1 / 12
About This Presentation
Title:

Chapters 8

Description:

Design affected by many factors (expandability, resilience to failure, ... processor-memory (short high speed, custom design to maximize mem-processor bandwidth) ... – PowerPoint PPT presentation

Number of Views:65
Avg rating:3.0/5.0
Slides: 13
Provided by: toda98
Category:
Tags: chapters

less

Transcript and Presenter's Notes

Title: Chapters 8


1
Chapters 8
  • (partial coverage)

2
Interfacing Processors and Peripherals
  • Processors Memory focus on performance and cost
  • I/O focus on dependability and cost
  • Design affected by many factors (expandability,
    resilience to failure, performance)
  • Performance, smaller role, but more complex
    access latency throughput connection
    between devices and the system the memory
    hierarchy the operating system

3
I/O
  • Important but neglected The difficulties in
    assessing and designing I/O systems have often
    relegated I/O to second class status courses
    in every aspect of computing, from programming
    to computer architecture often ignore I/O or
    give it scanty coverage textbooks leave the
    subject to near the end, making it easier for
    students and instructors to skip it!
  • GUILTY! we wont be looking at I/O in much
    detail be sure and read Chapter 8 in its
    entirety. you should probably take a
    networking class!

4
I/O Devices
  • Very diverse devices behavior (i.e., input vs.
    output, storage) partner (who is at the other
    end? Human or machine) data rate (peak rate
    data is transferred between I/O and main
    memory or processor)
  • Ex Keyboard
  • Input device
  • Used by human
  • Peak Data Rate 10 bytes/sec

5
I/O Example Disk Drives
  • To access data seek position head over the
    proper track (3 to 14 ms. avg.) rotational
    latency wait for desired sector (.5 / RPM)
    transfer grab the data (one or more sectors)
    30 to 80 MB/sec

In 2004
6
I/O Example Buses
  • Shared communication link
  • One or more wires, used to connect multiple
    subsystems
  • Advantages
  • Versatility (Easy to add new devices)
  • Low cost (Single set of wires shared in multiple
    ways)
  • Disadvantages
  • Can be major bottleneck
  • Limiting the max I/O throughput
  • Difficult to design because of physical limits
  • Length of the bus
  • Number of devices
  • Different device latencies and data transfer rates

7
I/O Example Buses
  • Consist of
  • Set of control lines
  • Send request and acks (indicate type of info on
    data lines)
  • Set of data lines
  • Between the source and the destination (data,
    complex commands, addresses)
  • Buses are shared resource
  • Need protocol to decide who uses it next
  • Bus transaction
  • 2 parts sending address and receiving or sending
    data
  • Input and output transactions
  • Input inputting data from the device to memory
    where the processor can read it
  • Output outputting data to a device from memory
    where the processor wrote it

8
I/O Example Buses
  • Types of buses
  • processor-memory (short high speed, custom design
    to maximize mem-processor bandwidth)
  • I/O (lengthy, different devices, e.g., USB,
    Firewire, wide range bandwidth of devices)
  • Backplane interface connection between the I/O
    and memory buses (high speed, often standardized,
    e.g., PCI)
  • Others buses do exist, such as graphics buses
  • Types of Communication protocol
  • Synchronous
  • Clock in control lines
  • Fixed protocol for communication relative to the
    clock
  • Fast and small protocol (implement in FSM)
  • But every device must operate at same rate and
    clock skew requires the bus to be short
  • EX processor-memory bus
  • Asynchronous
  • Dont use a clock
  • Instead use handshaking
  • Accommodates wide range of devices
  • No clock skew or synchronization problems
  • Bus can be lengthy

9
I/O Bus Standards
  • Today we have two dominant bus standards

10
Pentium 4
  • I/O Options

11
Other important issues
  • Bus Arbitration daisy chain arbitration (not
    very fair) centralized arbitration (requires
    an arbiter), e.g., PCI collision detection,
    e.g., Ethernet
  • Operating system polling interrupts
    direct memory access (DMA)
  • Performance Analysis techniques queuing
    theory simulation analysis, i.e., find the
    weakest link (see I/O System Design)

12
Fallacies and Pitfalls
  • Fallacy the rated mean time to failure of disks
    is 1,200,000 hours (140 years), so disks
    practically never fail.
  • Marketing numbers, measured by running disk
    simultaneously and counting failed ones. Really,
    3.6 of disks will fail over 5-year period
  • Fallacy magnetic disk storage is on its last
    legs, will be replaced.
  • Reliability, no volatility, low cost, reasonable
    access time
  • Might become true with solid state
  • Fallacy A 100 MB/sec bus can transfer 100
    MB/sec.
  • Cant use 100 of resource, at best 70-80
  • 1MB of storage ? 1 MB/second bus speed (base 2
    vs. base 10)
Write a Comment
User Comments (0)
About PowerShow.com