Business Information Distributed Data Architecture

1
Business Information Distributed Data Architecture

• IS306 Chap 2-3

2
Learning Objectives
Chap 2

• To design, analyze and evaluate networks, it is
  important to understand types of information and
  their business requirements.
• Todays networks are largely designed around a
  particular form of data communications
  architecture.

3
Business Information

• All business information exists in two forms
• Digital is represented as a sequence of
  discrete symbols (e.g., characters on a terminal
  keyboard)
• Analog always means a continuous form (e.g.,
  electrical signal from a microphone)
• In digital system, the information rate and the
  capacity of the channel is measured in bits per
  second (bps).
• In analog system, the information and channel
  bandwidth is measured in Hertz (Hz).
• Units like kilo, mega, and giga are commonly
  used.
• Infomation rate
• bps capacity
• Hertz (Hz) bandwidth

4
Analog Information

• Continuous signal
• Expressed as an oscillation (sine wave format) of
  frequency
• Example Analog electrical signal generated by a
  microphone in response to continuous changes in
  air pressure that make up sounds
• Variables
• Frequency
• of cycles/sec Hz
• 1 cycle/sec 1 Hz
• Amplitude
• Phase

5
Basic Analog Terms

• Wave frequency The number of times a cycle
  occurs in given time period
• Wave amplitude Height of a wave cycle
• Hertz (Hz) The number of times a wave cycle
  occurs in one second (commonly used measure of
  frequency)

6
Analog Signaling
1 cycle
Phase difference
Amplitude (volts)
time
(sec)
Frequency (hertz)
cycles per second
7
Analog Signaling

• ? the distance between the midpoints of the wave
  (sine wave)
• Wavelength ? has an inverse relationship to
  frequency f.
• c the speed of wave, or velocity of wave ( )
• When dealing with electromagnetic radiation in a
  vacuum, this speed is the speed of light. For
  sound waves in air, this is the speed of sound.
• Light c 299,792,458 m/s (1,079,252,848.8 km/h)
• Sound c 344 m/s (1,238.4 km/h)

8
Units (comm. vs. storage)
9
Digital Information

• Represented as a sequence of discrete symbols
  from a finite alphabet of text and/or digits
• Rate and capacity of a digital channel are
  measured in bits per second (bps)
• Digital data is binary uses 1s and 0s to
  represent everything
• Binary digits can be represented as voltage pulses

10
Basic Digital Terms

• Bit digit in a binary number
• 1 is a 1-bit number (1 in base 10)
• 10 is a 2-bit number (2 in base 10)
• 10011001 is an 8-bit number (153 in base 10)
• Byte eight bits

11
Types of Information

• Information sources can produce information in
  analog or digital form.
• Most information networks are designed to provide
  one of the following four services
• Audio
• Data
• Image
• Video

12
Voice Information

• Human voice is analog in nature. We can produce
  sounds in the range of 20 Hz to 20 KHz.
• But the quality of sound is dependent upon the
  channel bandwidth.
• Telephone network only passes signals between 300
  Hz to 3,400 Hz. (e.g., Bandwidth 3,400 300
  3,100 Hz)
• Teleconferencing audio requires up to 7,000 Hz.
• Hi-Fi stereo requires 15,000 Hz, while CD quality
  audio requires 20,000 Hz per stereo channel.
• Voice can be digitized. Any analog signal can be
  digitized.

13
Sampling Encoding

• Nyquist criterion sample the analog signal at a
  rate at least equal to twice the maximum
  frequency
• For a phone line with 4000 Hz max frequency
  passed, sampling rate must be 8000 samples/sec.
• Encode each sample with number of bits. 8
  bits/sample is typically used.
• The digital rate generated is 8 x 8,000 64,000
  bps.
• Human speech has an average duration of 1 to 5
  mins with intervals of talk and silent periods
  making it bursty.

14
Voice Sampling Example
4,000 Hz

• Sampling 5.2, 5.7, 6.1,
  3.4, .
• Quantization 5, 6,
  6, 3, .
• Encoding 00000101, 00000110,
  00000110, 00000011,
• DS-0 64 kbps (or 56 kbps typical channel
  capacity)
• DS-1 (T-1) 1.544 Mbps
• DS-3 44.736 Mbps

10

• Sampling rate
• 2 max freq.
• 8,000 smp/s
• 8 bits to each sample
• Digitized voice rate
• 8,000 smp/s x 8 b/smp
• 64,000 bps
• 64 kbps ? DS-0 (64 kbps)

0
1 DS-0 2 .. 24 T-1 (DS-1)
64 kbps
15
Data

• Data is a digital information to begin with and
  is typically generated by computers.
• An 8.5 in x 11 in page containing 1 in margins
  all sides and text which is double-spaced (3
  lines/in) and 10 pt type face (10 chars/in)
  requires how many bits.
• 8.5 - 2 6.5 x 10 65
• 11 - 2 9 x 3 27 ? 65 x 27 1,755 chars
• 1755 x 8 bits 14040 bits 10,000 bits
• Time to transmit
• size of data/data rate 10,000 bits/56,000 bps
  0.18 sec
• Total delay
• TxTime preparation delay 0.18 0.5 0.68
  sec
• Data applications range from email, ftp, web
  browsing to database query processing and
  real-time transactions.
• Response time becomes critical in the network
  design

16
Image Information

• Several applications create and use images
• Facsimile text with graphics, signature
• CAD/CAM computer graphics
• Publishing bmp, gif, tiff, jpeg
• Medical imaging MRI, X-Rays, CT-Scan

Pixel denotes light intensity 1 bit B/W, 8
bits shades of grey Color requires 3 frames for
R, G, and B
MRI image 256 x 256 x 12 bits 0.13 MB/image
50 images/exam (DS-0 928 sec/exam) CT-Scan 512
x 512 x 12 bits 0.52 MB/image 30 images/exam
(DS-0 2247 sec/exam) (Computed Tomography) Post
Script uses a different format DS-0 56 kbps,
DS-1(T-1) 1.544Mbps, DS-3 44.736 Mbps
17
Image Quality Issues

• More pixels better quality larger size
• More compression reduced quality increased
  speed
• Lossy compression can reduce the data by
  factors of roughly 101 to 201
• Lossless compression reduces the data less than
  51 (for medical imaging)
• Format (vector vs bitmapped/raster) affects size
  and therefore bandwidth requirements
• Choices in imaging technology, conversion, and
  communication all affect end-users satisfaction

18
Vector vs Raster

• Raster graphics
• Vector graphic

An image is represented as a two-dimensional
array of spots, called pixels.

• 1 bit/pixel 1-black 0-white
• 3-bit gray scale produces 8 shades of gray (w-b)
  
• if 8 b/pixel ? 256 shades of gray
• RGB each pixel is defined by three values,
• one for each of the three colors

pixel
Position description Postscript file language
description file
An image is represented as a collection of
straight and curved line segments. Simple objects
and more complex objects are defined by the
grouping of line segments
19
JPEG Image Compressions 1 to 99
JPEG Image 99Compression 1.64KB
JPEG Image 1Compression 44.6KB
JPEG Image 50Compression 14.2KB
20
Video Information

• The human eye has the property that when an image
  is flashed on the retina, it is retained for some
  number of milliseconds before decaying. If a
  sequence of images is flashed at 30 images/sec,
  the eye does not notice that it is looking at
  discrete images. All video (i.e., television)
  systems exploit this principle to produce moving
  pictures.

21
Standard TV Video

• NTSC video uses 525 scan lines, 43 aspect ratio
  and 30 frames/sec.
• PAL/SECAM video uses 625 scan lines, 43 aspect
  ratio and 25 frames/sec

scan line painted on the screen
Scan line
1
3
horizontal retrace
525
vertical retrace
22
Digital Video

• Total bandwidth 640 x 480 x 8 x 3 bits/color
  frame x 30 frames/sec 221 x 106 bps 221
  Mbps
• No wonder, we dont have video over our phone
  lines

640
Frame 640 x 480 (VGA) 8 bits/pixel (represent
256 gray scales) 30 frames/sec for motion video 3
RGB frames for color video
480
pixel
23
Response Time (RT)

• Response Time (RT)
• The time it takes a system to react to a given
  input. The time between the last keystroke by the
  user and the beginning of the display of a result
  by the computer
• User response time (UT)
• The time span between the moment a user receives
  a complete reply to one command and enters the
  next command (think time)
• System response time (ST)
• The time span between the moment the user enters
  a command and the moment a complete response is
  displayed on the terminal
• Network transfer time (NT)
• RT ST (include UT) NT
• RT lt 2 sec acceptable
• RT gt 15 sec intolerable

24
Bandwidth Requirements

• Review Fig. 2.7
• What happens when bandwidth is insufficient?
• How long does it take to become impatient?
• Is data communication ever fast enough?

25
Chapter objectives
Chap 3

• Describe the difference between centralized and
  distributed data processing and discuss the pros
  and cons of each approach.
• Explain why a distributed data processing system
  needs to be interconnected with some sort of data
  communications or networking facility.
• Describe the different forms of distributed data
  processing for applications.
• Describe the different forms of distributed
  databases.
• Discuss the implications of distributed data
  processing in terms of the requirements for data
  communications and networking facilities.
• Understand the motivation behind the trend to
  client/server architecture.

26
Centralized Data Processing

• Centralized computers
• One or more computers are located in a central
  facility.
• Centralized processing
• All applications are run on the central data
  processing facility.
• Centralized data
• Most data are stored in files and databases at
  the central facility and are controlled by and
  accessible by the central computer or computers.
• Centralized control
• A data processing or information systems manager
  has responsibility for the centralized data
  processing facility.
• Centralized support staff
• A centralized data processing facility must
  include a technical support staff to operate and
  maintain the data processing equipment.
• What are the advantages?
• Economies of scale (equipment and personnel)
• Lack of duplication
• Ease in enforcing standards for programming and
  data file structure, maintaining control over
  procurement, designing and implementing a
  security policy

27
Holiday Inn Centralized Architecture

• An example of a centralized data processing
  facility

workstation
Satellite Link
Leased Line
Lan
Lan
server
U.S. Hotel
servers
Mainframes
Corporate HQ
Lan
Satellite Link to Europe
Workstations
Corporate Data Center
28
Distributed Data Processing

• Computers are dispersed throughout an
  organization
• Allows greater flexibility in meeting individual
  needs
• More redundancy
• More autonomy
• Why is DDP increasing?
• Dramatically reduced workstation costs
• Improved user interfaces and desktop power
• Ability to share data across multiple servers
• DDP pros and cons
• There are no one-size-fits-all solutions
• Key issues
• How does it affect end-users?
• How does it affect management, productivity, and
  bottom-line?

29
Benefits of DDP

• Responsiveness
• Availability
• Correspondence to organizational patterns
• Resource sharing
• Incremental growth
• Increased user involvement and control
• Decentralized operation and centralized control
• End-user productivity
• Distance and location independence
• Privacy and security
• Vendor independence
• Flexibility

30
Drawbacks of DDP

• More difficult test and failure diagnosis
• More dependence on communications technology
• Incompatibility among equipment
• Incompatibility among data
• Network management and control
• Difficult to overall management, to enforce
  standards for software and data, and to control
  the information available through the network
• Difficulty in control of corporate information
  resources
• It may become difficult for a central authority
  to control the integrity and security of the data
• Suboptimization
• The totality of procurements throughout an
  organization may exceed the total requirement
• Duplication of effort

31
Reasons for DDP

• Need for new applications
• On large centralized systems, development can
  take years
• On small distributed systems, development can be
  component-based and very fast
• Need for short response time
• Centralized systems result in contention among
  users and processes
• Distributed systems provide dedicated resources

32
The DP Pendulum

• Centralized systems (mainframes, etc)
• Distributed systems (PCs)
• Networked systems
• Client-Server computing
• Client/Server Architecture
• Combines advantages of distributed and
  centralized computing
• Cost-effective and achieves economies of scale by
  centralizing support for specialized functions
• Flexible, scalable approach
• File service and database services can be on the
  same computer or be provided by several
  distributed machines

MF 60s
DDP 70-80s
Network Com. 80-90s
C/S 90s
P2P 00s
33
What is Client/Server?

• To us, client/server systems contain a client
  part and a server part
• The client makes requests for some service and
  the server responds by providing that service
• Most client/server systems are distributed i.e.,
  client part runs on one computer and the server
  runs on another computer
• Client/Server systems may not be distributed in
  this case both client and server run as separate
  processes on the same computer

34
Typical Client/Server Environment
PC
Print Server
X Terminal
Mainframe Database Server
Router
File Server
Mail Server
Name Server
LAN
Compute Server
Dumb Terminals
Mobile
35
Our Expectations

• As a technical professional or manager, what do
  you want to do with this diverse collection of
  equipment?
• Single System View access any systems power
  and resource and achieve seamless
  interoperability
• Integration restrict access to authorized
  clients only
• Reliability failures must have minimal impact
• Scalability
• What is available now and when can you have you
  needs met?

36
Single System Computing
Interface
User
Presentation logic
Logic
Computation logic
Information logic
Information storage and Retrieval
37
Functions of the Modules

• User Interface Module
• Contains all user interaction through displays,
  keyboards, mice etc. The display may be graphical
  (MS Windows) or character-based.
• Information Storage and Retrieval Module
• Reads data from and writes data to storage
  (disks, tapes, CD-ROMs)
• Logic Modules
• Perform all CPU-based processing normally called
  computation
• Presentation logic displays graphical output
  and processes user input
• Computation logic performs severe computation
  (simulation, statistics, economic forecasts)
• Information logic determines how to access and
  retrieve data

38
Logic Information Partition
Network
User
Presentation
Computation
Client
Information
Distributed DBMS
Distributed Data Objects
Information storage retrieval
Distributed File System (Novell Netware, NFS)
File transfer (FTP)
39
Partition Everything
X Server
Telecom. Network
User Interface
PC Client
Presentation
Compute Server
Computation
Information logic server
Information
Information storage retrieval
DBMS Server
40
Intranet

• Uses Intranet-based standards, such as HTML and
  SMTP (simple mail transfer protocol)
• Uses the TCP/IP protocol suite for local and wide
  area networking
• Content is accessible only to internal users
• Comprises wholly owned content not accessible to
  the public Internet even though the corporation
  has Internet connections and runs a Web server on
  the Internet
• Can be managed, unlike the Internet
• A specialized form of client/server architecture
• Advantages ease of implementation and use

41
Extranet

• Similar to Intranet, uses TCP/IP protocols and
  Web applications
• but provides access to controlled number of
  outside users, in a fashion that enforces a
  security policy
• Vendors/suppliers
• Customers
• The typical model of the extranet is
  client/server architecture

42
Network Implications

• Connectivity requirements
• What links between components are necessary?
• In a vertically partitioned DDP system,
  components of the system generally need links
  only to components above and below them in the
  hierarchical structure.
• In a horizontally partitioned DDP system, it may
  be necessary to allow data exchange between any
  two systems.
• Availability requirements
• The percentage of time that a particular
  function, application or data is available for
  users.
• High availability requirements mean that the
  communications facility must be highly available.
• Performance requirements
• e.g., response time requirements
• Depends on the nature of the DDP system and the
  applications