Computing Environments CSC8304

1
Computing EnvironmentsCSC8304

• Marcus Kaiserhttp//www.biological-networks.org
  

2
About this module

• Module Leader Dr Marcus Kaiser
• Todays lecture Introduction to Computing
  Environments
• Lectures for the next 4 weeks cover the need for
  database systems, statistical packages, security
  and poll-based/event-driven software
• Reading Week then follows no lecture
• Remaining lectures cover programming languages,
  scripting and Perl
• Lectures are on Tuesdays, 930-11am, CLT.602
• Lecture notes can be found on http//www.biologica
  l-networks.org/
• (Training gt Computing Environments for
  Bioinformatics )
• Practical classes start in DAYSH.821 and are on
  Wednesdays 11-12am

3
Assessment

• Coursework
• Databases/SQL 5 Nov deadline 15 of final mark
• Scripting/Perl 10 Dec deadline 15 of final
  mark
• Exam
• Exam January 70 of final mark

4
A computing environment
Enabling presentation of services to users and
interaction with such services by users
Monitors
Tactile feedback devices
Robots

• Requirements

A world of supporting services
Services
Mobile devices (e.g., mobile phones)
Video game consoles
Printers
5
A heterogeneous environment

• A modern day computing environment is made up
  from many different types of enabling
  technologies.
• An enabling technology refers to the mechanism
  that is used to implement a service.
• Many technologies share the same ultimate goal
  (e.g., sending a message from one computer to
  another).
• However, such technologies may attempt to achieve
  the same goal in different ways (e.g., Microsoft
  and Linux operating systems).

6
Standards

• There are instances when vendors must adhere to
  some standard to ensure integration (the Internet
  protocols exemplify this).
• Standards play a crucial role in computer system
  development.
• There are two types of standard
• Provided by an organisation that is recognised by
  the community as assuming the role of identifying
  standards (members of such an organisation are
  usually drawn from different vendors).
• Provided by a vendor (or group of vendors) and
  deployed without international recognition
  (however, such recognition may occur at a later
  date).

7
Computer technology evolution
1945 ENIAC http//en.wikipedia.org/wiki/Eniac 2004Pentium
Complexity 18,000 Valves X 103 42 M transistors
Size 200 m3 X 10 -8 6 cm3
Speed 150 ops/s X 106 1.6 x 109 ops/s
Consumption 10 kW X 10-3 68W
Cost 10 000 000 X 10-4 lt1000
Reliability Hours X 1000 Years
8
What if cars improved in a similar fashion?!!
Speed 70 mph X 106 11000km/s
Fuel 50 mpg X 10-3 50,000 mpg
Cost 10,000 X 10-4 1
Reliability 1 Year X 1000 1000 Years
Weight 1 ton X 10-8 10 mg
9
Conceptual Levels of Computers

• A digital computer is capable of computation,
  communication and storage.
• Computation is performed by carrying out
  instructions, a sequence of instructions to solve
  a problem is called a program
• Instructions are recognised and executed by the
  electronic circuits in the computer
• Instructions can only permit simple operations,
  typically, they can
• Add two numbers
• Check if a number is zero
• Move data around in Memory
• This set of instructions is called the machine
  language
• Different types of computers usually have
  different machine languages
• The machine language is said to be the interface
  between the software and the hardware.

10
Conceptual Levels of Computers contd.

• Most users do not use machine language
• Use high level language
• e.g. Java
• The high level language is translated to machine
  language by a compiler
• Computers can be thought of as having different
  levels, each with its own language.
• Each level carries out tasks on behalf of the
  level above it.
• Helps to cope with understanding the complexity
  of computing systems

Application Software (anybody)
Software
High Level Language (Java programmer)
Operating System Level (programmer)
Assembly Language Level (Assembly programmer)
Software and/or Hardware
Conventional machine level (hardware designer)
Integrated circuit level (VLSI designer)
Transistor level (Physical designer)
Hardware
Silicon electronics level Chemical engineer
11
Data Representation (1)

• Humans count in base 10, using 10 digits
• Difficult to represent electronically
• Machines count in base 2
• Two-state devices are easy to make (transistors)
• Only two digits used (0 and 1) called binary
  digits or bits
• Electrically represented by 0 volts and 5 volts
• Each bit occupies one memory cell or wire
• The basic working unit consists usually of 8
  bits, called a byte
• The basic memory unit is a multiple number of
  bytes e.g.
• 2 bytes 16 bits
• 4 bytes 32 bits
• 8 bytes 64 bits
• The basic memory unit is called the word length

12
Data Representation (2)

• All bytes in the memory are numbered, or
  addressable
• The first byte is numbered 0, the second 1, and
  so on
• Memory size is usually expressed in terms of
  (Mega)bytes
• It is common practice to
• Write the least significant bit (LSB) on the
  right
• Write the most significant bit (MSB) on the left
• Start counting from zero
• All data held within the computer is represented
  by a number of bits or bytes
• All high level objects, such as a Java or C
  class must be translated into bits

13
Data Representation (3)

• Data comes in many forms
• Booleans
• Characters (i.e. text)
• Integers, both positive and negative e.g. -230,
  -1, 0, 45319
• Real numbers, also called floating point numbers,
  e.g. 3.0, log (13), sin(p/4), 22/7
• Structured data types defined by programming
  languages e.g.
• Arrays
• Strings
• Classes
• Each type is represented by one or more bits

14
Bits, Bytes, and Buzzwords

• Terms used to describe file size or memory size

• Byte
• Kilobyte (KB)
• Megabytes (MB)
• Gigabytes (GB)
• Terabytes (TB)

• 8 bits
• 1024 (210) Bytes
• 220 Bytes
• 230, or about a billion, Bytes
• 240, or about a trillion, Bytes

15
Integer data

• Integer numbers dont allow fractions
• Humans use the decimal number system. There are
  10 digits, 0 9.
• Each place within a decimal number represents a
  power of 10. For example
• 236
• 2 102
• 3 101
• 6 100
• 10 is not a natural base (it is an anatomical
  incident!)
• Computers work more naturally with base 2 because
  transistors have two states
• In base 2, only digits 0 and 1 are used. Greatly
  simplifies the arithmetic and makes it much faster

16
Binary Numbers

• Each place within a binary number represents a
  power of 2.
• e.g binary 101
• 1 x 22
• 0 x 21
• 1 x 20
• (equals five in decimal)
• Electrical representation three wires

ON
OFF
ON
(5V)
(0V)
(5V)
17
Binary Arithmetic

• Humans perform decimal addition by
• Memorising all single-digit additions
• Writing the numbers to be added down
  right-aligned, one above the other
• Starting at the right and working towards the
  left
• Adding the digits, writing down the result and
  propagating any carry to the next column
• Subtraction works much the same way except that
  you must borrow from the next column
• Multiplication with a single-digit number works
  much the same way too
• Multiplication with a multi-digit number is
  treated as a series of separate single digit
  multiplications, the results of which are added
  together
• Binary addition, subtraction and multiplication
  can treated exactly the same except that only the
  digits 0 and 1 are used.

18
Basic Binary Arithmetic - examples
1 0 0 1 0 1 0
0 0 1 1 1 0 1
- - - - - - -
1 1 0 0 1 1 1
0 0 1 1 0 1 0
0 0 0 1 0 0 1 -
- - - - - - -
0 0 1 0 0 0 1
19
Hexadecimal Numbers

• Problem with binary arithmetic Long strings are
  fine for machines but awkward for humans
• e.g. What is the binary number 0100101011100011
  ??
• Guess then work it out!
• We (humans) therefore often use hexadecimal
  numbers (or hex for short).
• This uses base 16.
• There are 16 digits (0 1 2 3 4 5 6 7 8 9 A B C
  D E F)
• Each place represents a power of 16
• e.g. 29F 2 162 9 161 F 160
  (671 in decimal)

20
Integer Representation

• For the sake of economy, different hardware
  representations are used to implement different
  integer ranges
• The following are commonly found

Name Bits Range signed Range unsigned
Byte 8 -128 127 0 .. 255
Word 16 -32768 .. 32767 0 .. 65535
Long 32 -231 .. 231-1 0 .. 232-1
Quad 64 -263 .. 263-1 0 .. 264-1
21
Integer overflow

• It is possible that the result of an integer
  calculation is bigger than the allowed maximum
  (both positive and negative)
• Look at the following 8-bit addition
• 11001000 200 -56
• 10010110 150 -106
• -------------- ----- ------
• (1)01011110 (256) 94 (-256)
  94
• The final carry disappears because there is no
  hardware provision for it. The problem is called
  overflow (or underflow)
• Is this serious? Would you like this to happen to
  your bank account?
• Overflow is a serious problem. It indicates the
  presence of a bug in your program. The hardware
  can detect overflow and will cause your program
  to crash
• Overflow occurred in the European Space Agencys
  Ariane 5 rocket when the on-board software
  attempted to fit a 64 bit number into 16 bits.
  This did indeed cause the program to crash...

22
Floating Point data

• The range of possible values using 32 bits to
  represent a number, positive or negative, is
  large
• However, bigger number representations are
  needed.
• e.g. numbers to allow fractions and powers as
  required by many scientific applications
• To represent fractions using integers, you would
  need two of them
• One for the numerator and one for the denominator
• Would be a major nuisance not computationally
  amenable
• The way to do this is to use floating point
  numbers.
• Floating point data types allow a much greater
  range of possible values
• They are represented in floating point notation

23
Floating Point Notation

• Details of how floating point values are
  represented vary from one machine to another.
• The IEEE standard is one of the standard floating
  point representations
• More info at http//www.cs.uaf.edu/cs301/notes/Ch
  apter4/node13.html

24
Character Data

• Used for textual data, but can represent small
  integers
• Usually held in a byte although commonly only 7
  bits are needed
• There are two major character sets
• EBSIDIC (on IBM mainframe machines)
• ASCII (on all other machines)
• We concentrate on ASCII (American Standard Code
  for Information Interchange)
• It has been standardised by ISO (International
  Standardisation Organisation)
• ASCII was actually designed for use with
  teletypes and so the descriptions are somewhat
  obscure
• Often text documents are referred to as in
  ASCII format easier for document interchange

25
ASCII

• The characters are classed as
• Graphic characters (printable or displayable
  symbols)
• Control characters (intended to be used for
  various control functions, such as vertical
  motion and data communications.
• The basic ASCII set uses 7 bits for each
  character, giving it a total of 128 unique
  symbols.
• The extended ASCII character set uses 8 bits,
  which gives it an additional 128 characters.
• The extra characters represent characters from
  foreign languages and special symbols for drawing
  pictures.

More info _at_ http//www.jimprice.com/jim-asc.htm
26
Unicode

• Unicode is a new system to standardise character
  representation
• Unicode provides a unique number for every
  character, independent of platform, program, or
  language
• Adopted by such industry leaders as Apple, HP,
  IBM, JustSystem, Microsoft, Oracle, SAP, Sun,
  Sybase, Unisys.
• Required by modern standards such as XML, Java,
  ECMAScript (JavaScript), LDAP, CORBA 3.0, WML
• An implementation of the ISO/IEC 10646 standard
• Enables Internationalization

27
Unicode

• How Unicode Works
• It defines a large (and steadily growing) number
  of characters (gt 110,000).
• Each character gets a name and a number, e.g.
  LATIN CAPITAL LETTER A is 65 and TIBETAN SYLLABLE
  OM is 3840.
• Includes a table of useful character properties
  such as "this is lower case" or "this is a
  number" or "this is a punctuation mark".
• The Unicode standard also includes a large volume
  of helpful rules and explanations about how to
  display these characters properly, do
  line-breaking and hyphenation and sorting
• Unicode is important do some extra reading!
  try a Google search

28
Summary

• A modern day computing environment is made up
  from many different types of enabling
  technologies
• Standards are used to permit interoperability
• Computers can be thought of as a number of
  different levels, ranging from the application
  software that we all use, right through to the
  electronic circuits within the computer
• Computers count in binary
• Various ways of representing numeric and
  character data