Computing Overview

Aspects of Computing

Software Life Cycle

Cooking Analogy

Algorithm Complexity

  • Performance measure using big O notation
  • Linear time algorithm: O(n) - running time linear in problem size
  • Quadratic time algorithm: O(n2) - running time is square of problem size
  • NP-Complete: so far, no polynomial time algorithm is known for the problem
  • Exponential time algorithm: O(en) running time exponential in problem size

Exercise 1:

  1. Given a set of numbers, describe an algorithm to find the minimum.
  2. What is the time complexity of your algorithm?

The field of computer science

  • Only a few decades old (unlike biology).
  • Current professions:
    • Computer Scientist (BS, MS, PhD)
    • Software Engineer (BS, MS)
    • Computer Engineer (BS, MS)
    • Information Technology Specialist (BS, MS, MBA)
    • Research/Academia (MS, PhD)
  • Major areas:
    • Theory/Algorithms
    • Systems (networks, OS, architecture, distributed computing)
    • Embedded systems
    • Software engineering
    • Artificial intelligence (including robotics, natural languages)
    • Graphics, human-computer interfaces
    • Languages (language design, compilers)
    • Scientific computing
    • Bioinformatics, computational biology

Computing Systems

Components of a Computing System:

  • Hardware
  • Operating System (OS)
  • Support/development tools (e.g. compiler, editor)
  • Applications

Hardware:

  • Von Neumann Architecture: data and programs stored in memory to be accessed by processor.

  • Memory

  • Processor

  • I/O:
    • Input: keyboard, mouse
    • Output: screen, printer
    • Input and output: floppy, CD/RW, zip-drive
  • More sophisticated systems:
    • Multiple caches (Memory bottleneck)
    • Faster processors (instruction-level parallelism)
    • Multiple processors
    • Parallel/super-computers
    • Distributed computing

Operating System

  • Collection of software as layer between applications and hardware

  • Control program that manages system resources

  • Examples: Unix, Windows, Linux, MacOS,...

  • Operating system concepts:
    • Process:
      • An executing program needs resources: memory, CPU, disk space
      • Process: a "sandbox" for a program to execute in.
      • OS manages/schedules multiple processes
    • API (Application Programmer Interface):
      • Mechanism for requesting resources from OS
      • Programmers usually do this via a library in a higher-level language

  • Goals of an OS:
    • Make hardware easy to use.
    • Efficient use of resources.
    • Facilitate multiple users, programs.
    • Standardize communication between programs.

  • Well-known OS's:
    • UNIX:
      • Created at Bell Labs in 70's.
      • Many "flavors" of Unix: System V, BSD Unix (Berkeley), Solaris (SUN).
      • Small variations among Unix flavors.
    • Linux:
      • Free Unix-type OS created by Linus Torvalds and programmers around the world.
      • Originally for PC's, but now also for servers.
      • Distributions: Redhat, Debian, Mandrake,...
    • Windows:

System software

  • Usually packed/sold with OS
  • Independent of applications and common to all applications
  • Examples: compilers, interpreters, libraries, user interface, database, resource management
  • Needed for basic running of the computer

Programming (Low Level)

  • Machine code: low-level instructions understood by hardware
  • Assembly code: "human-readable form of machine code"
    Slightly higher-level than machine code, usually 1-1 correspondence.
  • Machine/assembly code is Processor dependent
  • Initially, programs were written in machine code
  • High level programming languages: more efficient, readable and portable.
  • Consider the statement A = A + B:
    It can be replaced by a sequence of instructions in assembly

Exercise 2: Write assembly (low-level) code for A = B + C - A, by drawing a picture similar to the one above.

High-level programming languages

  • Why? Easier to understand, take less time to develop, portable.
  • Consider this simple HelloWorld example in Java: 

    
public class Helloworld {
  public static void main (String[] argv)
  {
    System.out.println ("Hello World!");
  }
}

  • Note: This program simply prints Hello World! to the screen.
  • It is a 6-line program.
  • Contrast: When converted into low-level code, it is over 400 bytes.

Compilation and Interpretation:

  • A compiler translates from high-level code to machine (low-level) code.
  • Compilation: only translation and not execution; execution is separate.
  • An interpreter translates and executes line-by-line.
  • Interpretation is usually slower for overall execution.
  • Compilation can optimize low-level code based on program structure.

Languages:

  • FORTRAN (1960's): one of the earliest high-level compiled languages.
  • C: (1970's) with UNIX, compiled language
  • C++: (1980's) based on C, compiled object-oriented language
  • Java:(1990's) syntax based on C++, interpreted object-oriented language
  • Interpreted (scripting) languages: Perl, Python,...