ZeePedia

Physical Media, Shared Medium, Switched Medium, Network Topologies, Seven-layer OSI Model

<< Difference between distributed computing and computer networks
img
Advanced Computer Architecture-CS501
________________________________________________________
Advanced Computer Architecture
Lecture No. 44
Reading Material
Patterson, D.A. and Hennessy, J.L.
Chapter 8
Computer Architecture- A Quantitative Approach
Summary
·
Physical Media (Continued)
·
Shared Medium
·
Switched Medium
·
Connection Oriented vs. Connectionless Communication
·
Network Topologies
·
Seven-layer OSI Model
·
Internet and Packet Switching
·
Fragmentation
·
Routing
Modem
To interconnect different computers by using twisted pair copper wire, an interface is
used which is called modem. Modem stands for modulation/demodulation. Modems are
very useful to utilize the telephone network (i.e. 4 KHz bandwidth) for data and voice
transmission.
Quality of Telephone Line
Data transfer rate depends upon the quality of telephone line. If telephone line is of fine
quality, then data transfer rate will be sufficiently high. If the phone line is noisy then
data transfer rate will be decreased.
Classification of Fiber Optic Cables
Fiber optic cables can be classified into the following types.
Multimode fiber
This fiber has large diameter. When light is injected, it disperses, so the effective data
rate decreases.
Mono mode Fiber
Its diameter is very small. So dispersion is small and data rate is very high.
Wavelength ­Division Multiplexing (WDM)
Waves of different wavelengths are simultaneously sent through fiber. So as a result,
throughput increases.
Wireless Transmission
Page 378
Last Modified: 01-Nov-06
Advanced Computer Architecture-CS501
________________________________________________________
This is another effective medium for data transfer. Data is transferred in the form of
electromagnetic waves. It has the following features:
·
Data rate is in Mbits/Sec.
·
Very effective because of flexibility.
·
Band width is much less than fiber.
Example 1
Suppose we have 20 magnetic tapes, each containing 40GB. Assume that there are
enough tape readers to keep any network busy. How long will it take to transmit the data
over a distance of 5Km? The choices are category 5 twisted-pair wires at 100Mbits/sec,
multimode fiber at 1500Mbits/sec and single-mode fiber at 3000Mbits/sec. (Adapted
from CA3: H&P)
Solution
The total amount of data
= total no. of mag. tapes x capacity of each tape
= 20 x 40GB= 800GB
The time for each medium:
Twisted pair = 800GB/100Mbits/sec
= 65536 sec = 18.2 hr
Multimode Fiber = 800GB/1500Mbits/sec
= 4369.06sec = 1.213 hr
Single mode Fiber = 800GB/3000Mbits/sec
= 2184.55sec
= 0.66hr
Car = time to load car + transport time + time to unload car
= 250sec + 5Km/30Kph + 250sec
= 500.16 sec = 0.13hr
Shared/Switched Medium
Shared Medium
If a number of computers are connected with a single physical medium (i.e. coaxial or
fiber), this situation is called shared medium. Because of many computers, collision takes
place and affects the data transfer rate. As the number of machines on a physical medium
increases, the data transfer rate decreases.
Switched Medium
To increase the throughput, a switched medium is used.
Example 2
Page 379
Last Modified: 01-Nov-06
img
Advanced Computer Architecture-CS501
________________________________________________________
Compare 20 nodes connected in three different ways: a single 100Mbits/sec shared
medium; a switch connected via cat5, each segment running at 100Mbits/sec; and a
switch connected via optical fiber, each segment running at 1500Mbits/sec. The shared
medium is 700m long, and the average length of each segment to a switch is 55m. Both
switches can support full bandwidth. Assume each switch adds 6µsec to the latency, and
the average message size is 200bytes. Ignore the overhead of sending or receiving a
message and contention for the network.
Solution
First we will calculate the aggregate bandwidth:
For shared medium
Aggregate bandwidth = 100Mbits/sec
For switched twisted pair
Aggregate bandwidth = 20 x 100Mbits/sec
= 2000Mbits/sec
For switched optical fiber
Aggregate bandwidth = 20 x 1500Mbit/sec
= 30,000Mbits/sec
Transport time = Time of flight + (message size/BW)
(700/1000)Km
Transport time shared = ---------------------- x 106µsec
(2/3 x 300,000)Km
+ (200 x 8bits / 100Mbits/sec)
= 3.5µsec + 16µsec = 19.5µsec
For the switches, the distance is twice the average segment. We must also add latency for
the switch.
(55/1000)Km
Transport time switch = 2x ---------------------- x 106µs
(2/3 x 300,000)Km
+ 6µsec
+ (200 x 8bits / 100Mbits/sec)
= 0.55µsec + 6µsec +16µsec
= 22.55µsec
Page 380
Last Modified: 01-Nov-06
Advanced Computer Architecture-CS501
________________________________________________________
(55/1000)Km
= 2x ---------------------- x 106µs
Transport time fiber
(2/3 x 300,000)Km
+ 6µsec
+ (200 x 8bits / 1500Mbits/sec)
= 0.55µsec + 6µsec +1.06µsec
= 7.61µsec
Although the bandwidth of the switch is many times that of the shared medium, the
latency for unloaded networks is comparable.
Connection Oriented vs. Connection less Communication
Connection Oriented Communication
·  In this method, same path is always taken for the transfer of messages.
·  It reserves the bandwidth until the transfer is complete. So no other server could
use that path until it becomes free.
·  Telephone exchange and circuit switching is the example of connection oriented
communication.
Connection less Communication
·  Here message is divided into packets with each packet having destination address.
·  Each packet can take different path and reach the destination from any route by
looking at its address.
·  Postal system and packet switching are examples of connection less
communication.
Network Topologies
Computers in a network can be connected together in different ways. The following three
topologies are commonly used:
·  Bus topology
·  Star topology
·  Ring topology
Bus Topology
In this arrangement, computers are connected via a single shared physical medium.
Star topology
Computers are connected through a hub. All messages are broad cast because the hub is
not an intelligent device.
Ring Topology
All computers are connected through a ring. Only one computer can transmit data at one
time, having a pass called "Token".
Page 381
Last Modified: 01-Nov-06
Advanced Computer Architecture-CS501
________________________________________________________
Seven Layer OSI Model
There are seven layers in this model.
1. Physical Layer
2. Data Layer
3. Network Layer
4. Transport Layer
5. Session Layer
6. Presentation Layer
7. Application Layer
OSI Model Characteristics
·  An interface is present between any two layers.
·  A layer may use the data present in another layer.
·  Each layer is abstracted from other layers.
·  The service provided by one layer can be used by the other layer.
·  Two layers can provide same service e.g. Check Sum calculated at different
layers.
·  On two machines, six layers are logically connected except the physical layer.
The physical layers of two machines are physically connected.
Internet and Packet Switching
Internet works on the concept of packet switching. Application layer passes data to the
lower layer and that lower layer passes data to the next lower layer and on so on. In this
data passing process through different layers, different headers are attached with the data
which shows the source and destination addresses, number of data bytes in packet, type
of message etc. At physical layer, this packet is transmitted into the network. At
reception, reverse procedure is adopted.
Fragmentation
When a packet is lost in the network, it is re-transmitted. If the size of the packet is large
then retransmission of packet is wastage of resources and it also increases the delay in the
network. To minimize this delay, a large packet is divided into small fragments. Each
fragment contains a separate header having destination address and fragment number.
This fragmentation effectively reduces the queuing delay. At destination, these fragments
are re-assembled and data is sent to the application layer.
Routing
Routing works on store-and-forward policy. There are three methods used for routing:
·  Source-based routing
·  Virtual Circuit
·  Destination-based routing
TCP/IP
Internet uses TCP/IP protocol. In the TCP/IP model, session and presentation layers are
not present, so Store-Forward routing is used.
Page 382
Last Modified: 01-Nov-06
Advanced Computer Architecture-CS501
________________________________________________________
Page 383
Last Modified: 01-Nov-06
Table of Contents:
  1. Computer Architecture, Organization and Design
  2. Foundations of Computer Architecture, RISC and CISC
  3. Measures of Performance SRC Features and Instruction Formats
  4. ISA, Instruction Formats, Coding and Hand Assembly
  5. Reverse Assembly, SRC in the form of RTL
  6. RTL to Describe the SRC, Register Transfer using Digital Logic Circuits
  7. Thinking Process for ISA Design
  8. Introduction to the ISA of the FALCON-A and Examples
  9. Behavioral Register Transfer Language for FALCON-A, The EAGLE
  10. The FALCON-E, Instruction Set Architecture Comparison
  11. CISC microprocessor:The Motorola MC68000, RISC Architecture:The SPARC
  12. Design Process, Uni-Bus implementation for the SRC, Structural RTL for the SRC instructions
  13. Structural RTL Description of the SRC and FALCON-A
  14. External FALCON-A CPU Interface
  15. Logic Design for the Uni-bus SRC, Control Signals Generation in SRC
  16. Control Unit, 2-Bus Implementation of the SRC Data Path
  17. 3-bus implementation for the SRC, Machine Exceptions, Reset
  18. SRC Exception Processing Mechanism, Pipelining, Pipeline Design
  19. Adapting SRC instructions for Pipelined, Control Signals
  20. SRC, RTL, Data Dependence Distance, Forwarding, Compiler Solution to Hazards
  21. Data Forwarding Hardware, Superscalar, VLIW Architecture
  22. Microprogramming, General Microcoded Controller, Horizontal and Vertical Schemes
  23. I/O Subsystems, Components, Memory Mapped vs Isolated, Serial and Parallel Transfers
  24. Designing Parallel Input Output Ports, SAD, NUXI, Address Decoder , Delay Interval
  25. Designing a Parallel Input Port, Memory Mapped Input Output Ports, wrap around, Data Bus Multiplexing
  26. Programmed Input Output for FALCON-A and SRC
  27. Programmed Input Output Driver for SRC, Input Output
  28. Comparison of Interrupt driven Input Output and Polling
  29. Preparing source files for FALSIM, FALCON-A assembly language techniques
  30. Nested Interrupts, Interrupt Mask, DMA
  31. Direct Memory Access - DMA
  32. Semiconductor Memory vs Hard Disk, Mechanical Delays and Flash Memory
  33. Hard Drive Technologies
  34. Arithmetic Logic Shift Unit - ALSU, Radix Conversion, Fixed Point Numbers
  35. Overflow, Implementations of the adder, Unsigned and Signed Multiplication
  36. NxN Crossbar Design for Barrel Rotator, IEEE Floating-Point, Addition, Subtraction, Multiplication, Division
  37. CPU to Memory Interface, Static RAM, One two Dimensional Memory Cells, Matrix and Tree Decoders
  38. Memory Modules, Read Only Memory, ROM, Cache
  39. Cache Organization and Functions, Cache Controller Logic, Cache Strategies
  40. Virtual Memory Organization
  41. DRAM, Pipelining, Pre-charging and Parallelism, Hit Rate and Miss Rate, Access Time, Cache
  42. Performance of I/O Subsystems, Server Utilization, Asynchronous I/O and operating system
  43. Difference between distributed computing and computer networks
  44. Physical Media, Shared Medium, Switched Medium, Network Topologies, Seven-layer OSI Model