Transcript Slide 1

9691 Computing
Paper 3
Section 3.1
The functions of Operating Systems
Functions of Operating Systems (OS)
 All Operating Systems (OS) have three main functions:
 Controlling and Managing Hardware
 Providing An Interface (Human-Machine and Machine-Software)
 Facilitating Application Software To Run
 Operating systems must:
 Provide and manage hardware resources
 Provide an interface between the user and the machine
 Provide an interface between application software and the
machine
 Provide security for data on the system
 Provide utility software to allow maintenance to be done
VCN ICT Department 2013
2
Features of OS
1. User interface: the two main types of user interfaces
are command line and graphical user interface.
2. Device drivers (Routines which control hardware).
Example: Keyboard driver
3. Multitasking capability which enables the computer
to run more than one program at one time.
(Example: Running a word processing program, spreadsheet
program and a database program at the same time in the
computer.)
4. Spooling: directs jobs and data files to a queue on a
backing store before sending them to their intended
peripheral device.
VCN ICT Department 2013
3
Features of OS (continued)
5. Security: ensures that users can keep their files
confidential
6. Scheduler: allocates job priorities and find and
resolve deadlocks using scheduling algorithms
7. Memory manager: allocates memory to jobs
and data
8. Interrupt handler: routine in the operating
system which puts interrupts in a queue until
they are processed by the processor.
9. Translators: converts source code of application
programs into object code or to machine code
VCN ICT Department 2013
4
Interrupts

An interrupt is a signal from a hardware device or an
instruction from a software program indicating the processor
the need for a change in execution.

The processor processes program instructions stored in the
memory. It fetches instructions one after the other and
executes them.

If a problem occurs (anywhere in the system) while the
processor is busy processing, the processor must respond to
the problem as soon as possible in order to avoid damage to
data and/or hardware.

This requires the processor to stop whatever it is doing and
give its attention to the problem.

The device (hardware) or software, where the problem has
occurred generates a signal, called an “interrupt” to get
processor’s attention.
VCN ICT Department 2013
This is a simple method a
computer uses to carry
out instructions.
5
Types of Interrupt
• I/O interrupt
– Generated by an I/O device to signal that a job is
complete or an error has occurred.
• E.g.
– Printer is out of paper or is not connected.
• Timer interrupt
– Generated at fixed intervals.
– Allows for display refresh and to control access to
processor in multi-access or multi-programming system.
Types of Interrupt
• Hardware interrupt
– E.g.
• Power failure which indicates that the OS must close
down as safely as possible.
• Program interrupt
– Generated due to an error in a program such as
violation of memory use (trying to use part of the
memory reserved by the OS for other use) or an
attempt to execute an invalid instruction (such as
division by zero).
Types of Interrupts (Summary)
• Hardware: Generated by hardware devices. (Hardware
failure. Hardware going off-line)
• Software: Generated by software applications.
(Violation of memory space. Virus found, division by
zero)
• Timer: Generated by internal clock. (Scheduling of
processes; an event is due)
• I/O: Generated by an I/O device.
(e.g Printer out of paper; CD copying completed)
VCN ICT Department 2013
8
Managing interrupts
• After execution of an
instruction, the
processor must see if an
interrupt has occurred.
• If yes, the OS services
that interrupt following
a new set of
instructions.
VCN ICT Department 2013
9
What happens if the current job is more
important than the interrupt?
• The Priority of the interrupt is compared with
current job.
– If higher:
• An appropriate interrupt service routine (ISR),
depending on the interrupt's importance, is loaded and
run.
• The Interrupt is serviced by the processor.
– If lower:
• The interrupt is placed in a queue.
• The current job continues with next cycle.
VCN ICT Department 2013
10
Resuming an interrupted program from where it
left off after interrupt has been Serviced
• The state of the current job is saved.
• This is done by saving the contents of all the registers
in the processor so that the OS can use them to
service the interrupt.
• The OS can load the register contents back in the
registers to resume the job being interrupted once
the interrupt has been serviced
VCN ICT Department 2013
11
What happens if several interrupts
occur at the same time?
• Place the interrupts in a queue and only allow
return to the originally interrupted program
when the queue is empty.
• However, if an interrupt is considered important
enough then an interrupt service routine (ISR) is
chosen, which will “mask” the interrupt, meaning
that any future interrupts are not considered
until the current interrupt is completed.
VCN ICT Department 2013
12
What happens if some interrupts are
more important than others?
• The Interrupt is
allocated a position
in the job queue
according to
priorities.
– How this priority is
decided is called
scheduling
VCN ICT Department 2013
13
Handling Interrupts - Summary
• The Current cycle is completed
• Then the priority of interrupt is compared with the
current job
• If higher:
– Contents of special registers are saved, the current job is
allocated a position in the job queue according to priorities
– Interrupt is serviced by OS.
– On completion, values of special registers from original
program are loaded and the original job is restored.
• If lower:
– Interrupt is allocated a position in the job queue according
to priorities.
– Current job continues with next cycle.
JOB SCHEDULING
• Scheduling is an OS process that starts and ends tasks
(programs), manages concurrently running processes, and
allocates system resources.
REASONS FOR SCHEDULING
 Maximize the use of the whole of the computer system
 Be fair to all users
 Provide a reasonable response time to all users, whether they
are on-line users or a batch processing user
 Prevent the system failing if it is becoming overloaded
 Make sure that the system is consistent by always giving similar
response times to similar activities from day to day
VCN ICT Department 2013
15
Input Output (I/O) bound Vs Processor
bound jobs
• I/O bound jobs are those that require relatively
little processing but do need to use the
peripheral devices substantially.
– E.g. Printing wage slips for the employees of a large
company
• Processor bound jobs are those that require a
large amount of processor time and very little use
of the various peripheral devices.
– E.g. Analysing the annual, world-wide sales of the
company which has a turnover of many billions of
Shillings.
VCN ICT Department 2013
16
Scheduling Example
• Job A is processor-bound and Job
B is I/O bound.
• I/O bound jobs are given more
time in order to allow them to
finish in reasonable time.
• Processor bound jobs are given
lesser time as they can do more
in less time and shouldn’t hold
the I/O bound jobs.
VCN ICT Department 2013
17
STRATEGIES FOR SCHEDULING
• Scheduling of processes requires careful considerations. The
following are the most common:
 Priority: Giving some jobs priority over others.
 I/O or Processor Bound: Deciding which job is processor bound and
which is I/O bound.
 Type of Job: Batch processing, on-line and real-time jobs all require
different response times.
 Resource Requirements: The amount of time needed to complete the
job, the memory required, I/O and processor time.
 Resources Used So Far: The amount of processor time used so far,
how much I/O used so far.
 Waiting Time: The time the job has been waiting to use the system.
VCN ICT Department 2013
18
Process handling
• A job can be in any of the
following states.
NOTE
– Ready [Ready to start]
– Running [on the system]
– Blocked [Waiting for a
peripheral]
• A job can only enter the running state from the ready state.
• The ready and blocked states are queues that may hold several jobs.
• On a standard single processor computer only one job can be in the
running state.
• All jobs entering the system normally enter via the ready state and
(normally) only leave the system from the running state.
VCN ICT Department 2013
19
Schedulers
• High Level Scheduler (HLS) is responsible for
placing a job in the ready queue. The HLS makes
sure that the system is not over loaded.
• Mid-Level Scheduler (MLS) is responsible for
swapping jobs between main and secondary
memory.
• Low-Level Scheduler (LLS) is responsible for
Moving jobs in and out of the ready state.
– The LLS decides the order in which jobs are to be
placed in the running state.
VCN ICT Department 2013
20
Scheduling policies
• There are many policies that may be used to
do scheduling, but they can all be placed in
one of two classes. These are pre-emptive and
non-pre-emptive policies.
• Pre-emptive policy: allows the LLS to put or
remove jobs from the RUNNING queue as and
when needed.
• Non Pre-emptive policy allows jobs to run
until they no longer need the processor.
VCN ICT Department 2013
21
Common scheduling algorithms
• First come first Served (FCFS): Simply means that the first job to enter the
ready queue is the first to enter the running state. This favors long jobs. (Non
pre-emptive by design)
• Shortest Job first (SJF): Simply means sort jobs in the ready queue in ascending
order of times expected to be needed by each job. New jobs are added to the
queue in such a way as to preserve this order. (Non pre-emptive)
• Round Robin (RR): This gives each job a maximum length of processor time
(called a time slice) after which the job is put at the back of the ready queue
and the job at the front of the queue is given use of the processor. If a job is
completed before the maximum time is up it leaves the system.
• Shortest Remaining Time (SRT): The ready queue is sorted on the amount of
expected time still required by a job. This scheme favours short jobs even more
than SJF. Also there is a danger of long jobs being prevented from running.
• Priority Queues (PQ): Priority Queues involve queues of different priorities.
Jobs in higher priority queues are executed first.
• Multi-level Feedback queues (MFQ): Involves several queues of different
priorities with jobs migrating downwards.
VCN ICT Department 2013
22
Other methods of allocating priorities:
All the policies discussed on previous slides are methods of
allocating priorities.
Here are some others:
• Amount of time already waited.
• Amount of processor time already given.
• Amount of peripheral time.
– Compares I/O jobs and asks “How much of a I/O job are
you?”
• Necessary response time.
– Compares real-time / on-line jobs and asks “How much of
a real-time / on-line jobs are you?” or “OK, you are a realtime / on-line job but how quickly do I need to respond to
you?”
Other methods of allocating priorities:
•
Importance of Job / Type of Job
–
Safety critical jobs will be given very high priority, on-line and real time
applications will also have to have high priorities.
• E.g.
1. A computer monitoring the temperature and pressure in a chemical
process whilst analysing results of readings taken over a period of
time must give the high priority to the control program.
– If the temperature or pressure goes out of a pre-defined range,
the control program must take over immediately.
2. A bank's computer is printing bank statements over night and
someone wishes to use a cash point, the cash point job must take
priority.
–
Also remember I/O bound jobs will also get priority as stated earlier.
–
Can be Pre-Emptive or Non-Pre-Emptive depending on if the queue is
checked after each cycle or not.
VCN ICT Department 2013
24
Memory Management
• Memory Management is the process of managing
jobs and their data in the memory.
 When a job needs to be processed, it is stored in
the main memory along with its data.
 In most of the modern OS, more than one job can
be loaded into the memory at one time.
 If several job are stored in the memory, their data
must be protected from the actions of other jobs.
VCN ICT Department 2013
25
Memory Mgt Scenario
MAIN MEMORY
Free (20 Kb)
Job D (30 Kb)
Job C (10 Kb)
Job B (20 Kb)
Job A (50 Kb)
Operating system
• Suppose that jobs A, B, C and D have been loaded in RAM by
the Mid level scheduler.
• Now, imagine that the job with the next highest Priority is
Job E and that is of Size 35Kb. It will wait because there’s
only 20Kb free.
• So the current jobs will run until one of them is completed.
• If it is job D, no problem because we shall have enough
space to run E.
 If it’s job A, still no problem because the space is enough.
 If it’s Job B, there’s a problem because the free gaps are not
enough, so
– we have to wait till there’s more free space. (Causing delays)
– Part of job E is put in the free space and the rest where B used to be.
(This causes fragmentation issues)
– An active job can be moved to another location creating adjacent free
space which is then used for the job (Makes heavy use of the
processor)
VCN ICT Department 2013
26
Virtual memory
• Virtual memory is the area on the disk that is used to store the pages of
jobs of currently running programs and data temporarily, which makes it
look like that the computer, has more memory than it actually has.
• Virtual memory is necessary to:
– allow programs to run that need more memory than is available;
– allow multiple programs to run which have combined memory
requirements more than the main memory available.
• Virtual memory will, in some way, split a program into blocks and load the
blocks that are needed, when they are needed – constantly swapping
blocks between main memory and the backing store.
• The switching between the virtual memory and main memory which
involves the disk continuously searching for pages is called disk thrashing.
• The two normal methods of splitting the programs into blocks are
segmentation and paging.
VCN ICT Department 2013
27
Paging
• In the paging memory-management
scheme, the data transferred between
virtual memory and physical memory is
split into equally sized blocks called
pages.
• An index keeps track of what is in each
page.
• Pages in virtual memory that are not
being executed are stored on disk in a
page file.
• A large job can span 2 or more pages.
VCN ICT Department 2013
28
Segmentation
• Segmentation avoids the possibility of splitting up code
halfway through because pages are not big enough.
• Segments are adjustable size blocks. I.e. the size of a
segment is not fixed whereas the size of a page is.
• A segment will hold an entire section of code, relying
on logical breaks in the code.
• Space is not wasted as it would be in a page if the page
size was bigger than the data going into it and the code
is not split as it would be if it was bigger than the page
size.
• Segments are also indexed to keep track of what is in
each page.
VCN ICT Department 2013
29
Spooling
• SPOOL (Simultaneous Peripheral Output On Line)
• It is a method used to place input and output on
a fast access storage device, such as a disk, so
that slow peripheral devices do not hold up the
processor.
• It allows for queues when several jobs want to
use peripheral devices at the same time
• It stops different input and outputs becoming
mixed up
VCN ICT Department 2013
30
Spooling (Cont’d)
Example
• If two or more jobs are sent to a
printer at the same time, the jobs are
sent to a spool queue where they
wait for the printer to be free.
• The spool queue only stores
reference to where the jobs are
stored on a hard drive.
• Spooling saves the user waiting for
the printer (or other slow device) and
allows the processor to do something
else while the printing (or other slow
process) finishes.
• (An interrupt would be sent to
request the next job from the spool
queue.)
Benefits of spooling
• avoids delays – printers are relatively
slow and so spooling frees up the
processor quickly and allows it to get
on with other jobs;
• Allows more than one print job to be
submitted at a time – each job will be
held in a queue and printed one at a
time – the use of a queue also allows
priorities to be set.
• in a multi-user system, provides a
method of keeping print-jobs
separate – it means that printouts
will not be muddled up.
• Lets the processor get on with
something else while the jobs are
queued
VCN ICT Department 2013
31
Modern Operating Systems
• There are two main types of OS:
– command line
– graphical user interface (GUI).
• Modern operating systems allow apparent
multi-tasking by switching between multiple
tasks very quickly.
• Each application gets a time-slice.
• When their processor time is up, and
interrupt occurs and control is passed to the
next application.
• If a program requests use of a hardware
device, the request is placed in a queue until
the device is available.
VCN ICT Department 2013
32
Booting process
• When the computer is first turned on, the CPU initializes itself.
• Part of this initialization is to look for the BIOS (Basic input/output
system).
• Some of the BIOS is stored in ROM (read only memory) but since it
is user-configurable, it is not completely stored in ROM, these
parts are stored in CMOS RAM.
– CMOS (complementary metal-oxide semiconductor) is a type of memory
chip with very low power requirements which uses a small battery to
retain data when the PC is turned off.
– The BIOS cannot be stored on the hard disk because it contains the code
to initialize such secondary storage devices.
• The BIOS will then run its power-on-self-test (POST).
VCN ICT Department 2013
33
Booting process
Power-on-self-test (POST)
• The computer runs the power-on-self-test (POST) routine that is stored in ROM. The POST
routine:
– Checks the BIOS chip and tests the CMOS RAM
– Verifies and clears CPU registers
– Checks hardware devices e.g. video card, secondary storage devices, keyboard, mouse
– Loads the address of the first instruction of the boot program into the program counter
(PC).
Boot Program
• The boot program (also known as the boot loader or the bootstrap ) is stored in the ROM. It
looks for an OS to load.
– On a typical PC, the OS will be loaded from the C drive but it will check the floppy drive
and CD drive beforehand. The order in which devices are checked for an OS is called the
boot sequence and can be configured in the CMOS setup.
• Once an OS has been located, the boot program will encounter the boot record, which tells
it where to find the beginning of the OS and the subsequent program that will initialize the
OS.
VCN ICT Department 2013
34
Booting process (Cont’d)
OS Initialization
• Once the program that initializes the OS is
loaded, the BIOS copies its files into memory
and the OS can take control over the rest of
the boot process.
• The OS performs another check of the
memory and loads the device drivers needed
by peripherals such as printers, scanners, mice
and keyboards.
VCN ICT Department 2013
35
File Allocation Table (FAT)
• The OS must be able to store and locate
files on a disk. It uses a File Allocation
Table (FAT) to do this.
• The FAT uses a linked list to point to the
blocks on the disk that contain files.
• To do this, the OS formats the disk by
dividing it into sectors and concentric
circles (called tracks).
• Two or more sectors on a single track
make up a cluster. The FAT is loaded into
RAM to speed up the search time and to
avoid continual disk access.
VCN ICT Department 2013
36
File Allocation Tables (Cont’d)
 This is an example of a file allocation table.
 Every cluster on the disk is listed. If the
cluster is empty, it has a zero entry.
 Otherwise, the cluster containing the
beginning of a file is labeled (with a
pointer) and links to the cluster containing
the next part of the file.
 The cluster containing the end of the file
has a null pointer.
We can see that cluster 1 is empty. Our first example file
1 begins in cluster 2, the next part is in cluster 3 and the
final part of the file is in cluster 5. File 2 begins in cluster
4, followed by cluster 6 with the remaining bits of the
file in cluster 7.
VCN ICT Department 2013
37
File Allocation Tables (Cont’d)
• The fragmentation of files is caused by deleting files.
 Note that a file is never deleted from the physical disk,
only the reference to it is removed from the FAT. Of
course, when the disk is full it will be overwritten.
• To find a file,
– the OS looks in the table for the filename and, if it finds it, gets the
cluster number for the start of the file. The OS can then follow the
pointers in the table to find the rest of the file.
• To delete a file,
– it just has to set the clusters it occupied to zero.
• To add a new file,
– the OS has to linearly search for clusters with zero entries to set
up the linked list.
VCN ICT Department 2013
38
END.
Visit this link for more info
• http://w2.compu2learn.co.uk/theory/advance
d-theory/331-the-function-of-operatingsystems/memory-management
VCN ICT Department 2013
39