Chapter I.

Introduction
1.1 What is an Operating System?
Definition
 A program that acts as an intermediary between a user of a computer and the computer

hardware.
 A system program which controls all the computer's resources and provides a base

upon which application programs can be written.

 A system program that controls execution of application programs.

Objectives
 Convenience – Top Down View – Virtual Machine

o Extending the real hardware and acting as a Virtual Machine

o Hiding the truth about the hardware from the user
o Providing the user a convenient interface which is easier to program
 Efficiency – Bottom Up View – Resource Manager
o Providing an orderly and controlled allocation of resources
o Improving resource utilization
 Ability to Evolve
o Adaptability to new changes
 Hardware upgrades/New types of hardware
 Need of new services
o Easy fixes of latent errors
Computer System Components
 Hardware - provides basic computing resources (CPU, memory, I/O devices).

 Operating system - controls and coordinates the use of the hardware among the various

application programs for the various users.

 Applications programs - define the ways in which the system resources are used to

solve the computing problems of the users (compilers, database systems, video games,
business programs).
 Users (people, machines, other computers).

1
1.2 Evolution of Operating Systems
 Studying evolution of Operating Systems is used to
o Understand key requirements of an Operating System
o Understand significance of the major features of modern Operating Systems
 Operating Systems and Computer Architecture are historically tied
o Architectural changes affect structure of Operating Systems
o Architectural changes affect performance of Operating Systems
 Major progress in Operating Systems
o Early systems (Wiring-up plug-boards, Serial Processing)
o Simple batch processing
o Spooling batch processing
o Multi-programmed batch processing
o Timesharing systems
o Personal computer systems
o Parallel systems
o Distributed systems
o Real time processing

2
i. Wiring-up Plug-boards
 Programming was done in machine language by wiring up plug-boards
 Plug-boards control the basic functions of the computer
 No Operating Systems, No programming languages, No operators
 Complex programming and machine underutilization
ii. Serial Processing
 The Technique
o Running a Job
 Programmer creates his program and punches it on cards
 He submits the card deck to an operator in the input room
 He reserves machine time on a sign-up sheet
 Operator set-up the job as scheduled
 Operator caries over the output to the output room
 Programmer will collect it later
o No Operating Systems
o No programming languages
 Features
o Users get access in series
o Writing programs was improved
 Drawbacks
o Wasted time due to scheduling and setup
o Wasted time while operators walking around the machine room
o Large machine/processor idle time
iii. Simple Batch Processing
 The Technique
o To run jobs
 Users submit jobs on cards
 Operator batches jobs together into a tape inside the input room
 Operator rewinds the input tape and brings it to machine room
 Monitor, a special system program in the system tape, runs each job
sequentially
 Output is written onto output tape and is brought to output room
 Output is printed offline in the output room
o Each job is structured using JCL (Job Control Language)
 To switch control back and forth between job and monitor
 To automate job set-up (loading source code, compilers, object code etc)
o Monitor controls sequence of events
3
 o Relies on some hardware features
 Memory protection
 Timer
 Processor modes
 Interrupts
o Operating Systems introduced for the first time
 FMS – Fortran Monitor System for IBM 701 and later for IBM 704
 IBMSYS – for IBM 7090/7094
 Features
o Move much of the work of the operator to the computer
o Improve machine utilization
 Handle scheduling problem
 Reduce set-up time
 Drawbacks
o Large turn-around time
o Large CPU idle time on heavily I/O bound jobs
o Large CPU idle time due to slower speed of I/O devices compared to CPU speed
o CPU idle time between batches of jobs: the computer still needs to read the deck
of cards before it can begin to execute the batch of jobs.
iv. Spooling Batch Processing
 The Technique
o Adds up Spooling technique to simple batch processing
 Spooling - Simultaneous Peripheral Operation Online
 It is the ability to read jobs/print job outputs while the processor executes
jobs
 Read jobs – from cards to disk
 Print job outputs – from disk to printer
 Used in input and output operations
 With spooling, whenever a running job finished, the OS could load a new
job from disk into the now-empty partition and run it
o Spooling is the first attempt of multiprogramming
 Features
o Avoids CPU idle time between batches of jobs
o Improve turn-around time: Output of a job was available as soon as the job
completed, rather than only after all jobs in the current cycle were finished.
 Drawbacks
o Large turn-around time
4
 o Large CPU idle time on heavily I/O bound jobs
v. Multi-programmed Batch Processing
 The Technique
o Multiprogramming
 Multiple jobs are loaded in a partitioned memory
 Processor runs the jobs by cycling through them in some order
 Processor works on each job for a specified period of time
o The system is responsible for
 Starting user jobs
 Spooling operations
 I/O for user jobs
 Switching between user jobs
 Ensuring proper protection
o Relies on some hardware features
 Memory protection
 Timer
 Processor modes
 Interrupts
 DMA (Direct Memory Access)
o OS/360 – the first multi-programmed batch system designed for IBM System/360
 Features
o Improve machine utilization
o Resemble a modern operating system
 Drawbacks
o Long response time: the time between submitting a job and getting back the
output was often several hours
o Poor interactivity between programmer and his program
vi. Timesharing/Time-slicing
 The Technique
o Variant of multiprogramming in which each user has an online terminal
o Processor time is shared among multiple users
o Its main objective is getting quick response time
o The computer provides fast, interactive service to a number of users
o CTSS – the first timesharing system developed for IBM 709 and later to IBM
7094
o CTSS supported 32 users at a time
 Features
5
 o Minimizes response time
o Support many users simultaneously
 Difference between multiprogramming and timesharing
o Multiprogramming
 Its objective is to maximize processor use
 JCL commands are source of directives to the OS
 Designed to keep the processor & I/O devices simultaneously busy
o Timesharing
 Its objective is to minimize response time
 Terminal commands are sources of directives to the OS
 Designed to be responsive to the needs of individual user
 Drawbacks of multiprogramming and timesharing
o Problems related with multiple programs in memory interfering with each other
o Problems related with multiple interactive users accessing the same file system
o Problems related with multiple programs competing for the same resource
vii. Personal computer systems
 Personal computers - computer system dedicated to a single user.
 I/O devices - keyboards, mice, display screens, small printers.
 User convenience and responsiveness.
 Can adopt technology developed for larger operating systems; often individuals have
sole use of computer and do not need advanced CPU utilization or protection features.
viii. Parallel systems
 Tightly coupled system - processors share memory and a clock; communication usually
takes place through the shared memory.
 Advantages of parallel systems:
 Increased throughput
 Economical
 Increased reliability
 Symmetric multiprocessing
 Each processor runs an identical copy of the operating system.
 Many processes can run at once without performance deterioration.
 Runs on a multiprocessor standalone computer system
 Processors share the same main memory and I/O devices and can perform
the same
 Jobs are scheduled across all processors
 Several jobs run in parallel
 Advantages
6
 o Performance: due to multiprocessing
o Availability: Failure of a single processor does not halt the machine
o Incremental growth: additional processors can be added easily
o Scaling: Range of products with different price and performance
 Asymmetric multiprocessing
 Each processor is assigned a specific task; master processor schedules and
allocates work to slave processors.
 More common in extremely large systems.
ix. Distributed systems
 Runs on a multi-computer system: collection of computers, each with its own memory,
storage device and other I/O modules
 Useful for sharing of the workload among the different computers
 Existence of multiple computers is transparent to the user: It appears as a
uniprocessor system
 It differs in critical ways from uniprocessor OSs
x. Real-time processing
 Programs are guaranteed to have an upper bound on tasks that they carry out
 Usually it is desired that the upper bound be very small
 Severely constrained by the timing requirements
 Usually used in dedicated/special purpose computers