Chapter Two

Processes and process management

By Abebaw S.
Outline
 Process and Thread
 The concept of multi-threading
 Inter process communication
 Race conditioning
 Critical Sections and mutual exclusion
 Process Scheduling
 Preemptive and non preemptive scheduling
 Scheduling policies
 Dead lock
 Deadlock prevention
 Deadlock detection
 Deadlock avoidance
 Process concept
 Process can be defined as
 Unit of work

 It is a program in execution.

 It is an instance of running program

 It is an entity that can be assigned to a processor

 A program is a set of instructions to perform a specific task.

 For example, when we write a program in C++ and compile

it, the compiler creates binary code. The original code and
binary code are both programs. When we actually run the
binary code, it becomes a process.
 Process (continue . . . )
 A single program can create many processes. For
example, when we open binary file multiple times,
multiple processes are created.
 Difference between program and process

Program Process
 Def: collection of  Def: an instance of an
instructions designed to running program.
accomplish a certain task.  Nature: active entity.
Nature: passive entity.  Lifespan: limited.
 Lifespan: much higher
 Process (continue . . . )
Differences (Continue …)
Program Process
 Resource: require  Resource: It requires CPU,
only memory space to memory address, disk,
store the instructions. Input/output during its
 Computation Time: lifetime.
has no computation  Computation Time: Takes
time and cost. a long time to access and
compute a single fact.
 Process (continue . . . )
 In main memory, a process has
four sections:
 Text: consists of the current

activity of a process represented

by:
 Value of the Program
Counter
 Contents processor’s
registers.
 Data: This has global and

static variables.
Fig: Process in Memory
Process (continue . . . )

 Stack: Contains temporary data such as function

parameters, return address , and local variables.
 Heap: Dynamically memory allocated to process
during its run time.
 Process State (Continue. . .)
 The process, from its creation to completion, passes
through various states.
 The state of the process defined by the current activity of
the process.
 Each process may be in one of the following states:
 New: The process is being created; Or a process which is

going to be picked up by the OS into the main memory.

 Ready: Processes that are in the main memory and are

prepared for execution; but not currently assigned to the

processor.
Process State ( Continue. . .)

Fig : Diagram of Process State

Process State ( Continue. . .)
 Running: Instructions are being executed. In this state
the process access the CPU.
 Waiting: The process is waiting for some event to

occur. Or when the process waits for a certain resource

(Example I/O) the OS moves this process to the
waiting state and assigns the CPU to the other process.
 Terminated: The process has finished execution.
 Process State ( Continue. . .)
Other states
 Suspended ready : A process in the ready state, which is
moved to the secondary memory from main memory due to
lack of resource.
 Suspended wait : If the process in waiting state, requires a
resources; but it lacks resources, then the OS removes that
process and put it in the secondary memory.
 Process State Transition
 A process transitions
 Are change of states of a process due to an
event.
 They are represented by arrows in a process state

diagram, shows the possible states and transitions of

a process.
 For example, a process can transition from ready to

running when the scheduler selects it for execution,

or from running to waiting when it requests an I/O
operation.
 The Fig in the next slide shows the process state
Process State Transition( Continue. . .)

Fig: Process State Transition

 Process State Transition
 Null to new: A new process is created for the implementation of a
process.
 New to ready: When the OS has allocated resources to it and it is
ready to be executed.
 Ready to running : When the CPU becomes available, the OS
selects a process from the ready queue depending on various
scheduling algorithms and moves it to the running state.
 Running to terminated: When a process completes its execution
or is terminated by the OS, it moves to the terminated state.
 Process State Transition( Continue. . .)
 Running to ready: This occurs when the running
process has achieved its maximum running time for
continuous execution.
 Running to Wait: If a process requests something
for which it is waiting, it is placed in the waiting state.
 Wait to Ready: A process in the waiting state is
moved to the ready state, when the event for which it
has been waiting occurs.
 Process Control Block (PCB)
 Each process is represented in the OS by PCB.
 PCB is a data structure that contains information of the
process related to it.
 PCB also known as a task control block.
 It contains many pieces’ of information associated with
specific process including:
 Process ID: Unique number given to each process by

operating system, used to identifying the process.

 Process state: The state may be new, ready, running,

waiting, and terminated etc.

 PCB (Continue …)

 Program counter: indicates the

address of the next instruction that
has to be executed for that particular
process.
 CPU register : The registers that are
being used by a particular process.
There are different kinds of registers
such as index registers, stack
pointers, and general purpose
registers etc.
Fig: PCB
 PCB (Continue …)
 CPU scheduling information: It has priority, scheduling
queue pointers (to indicate the order of execution), and
several other scheduling parameters.
 Memory-management information: represents the
memory that are being used by a particular process. It
include information such as the value of the base and limit
registers, the page tables, the segment tables etc.
 I/O Status Information: This information includes the
list of I/O device used by the process, and the list of files
etc.
 Threads
 A thread is the subset of a process
 Also known as the lightweight process.
 A process can have more than one thread, and these
threads are managed independently by the scheduler.
 All the threads within one process are interrelated to
each other.
 Threads have some common information, such as data
segment, code segment, files, etc. that is shared to
their peer threads.
 But it contains its own registers, stack, and counter.
Threads (Continue…)

Fig - Single-threaded and multithreaded processes

 Threads (Continue…)
Differences
Process Threads
 It is instance of a program  It is a segment of a process or
that is being executed. a lightweight process.
 Independent of each other  Threads are interdependent
and hence don't share a and share memory.
memory or other resources.  The OS takes all the user-
 Each process is treated as a level threads as a single
new by the operating system. process.
 If one process gets blocked  If any user-level thread gets
by the OS, then the other blocked, all of its peer
process can continue the threads also get blocked
execution.
 Threads (Continue…)
Process Threads
 Context switching between  Context switching between
two processes takes much time the threads is fast because
as they are heavy. they are very lightweight .
 The data and code segment of  Threads share data and code
each process are independent of segment with their peer
the other. threads;
 OS takes more time to  Threads can be terminated in
terminate a process. very little time.
 New process creation is more  A thread needs less time
time taking as each new process
for creation.
takes all the resources.
 Threads
 Advantages
 Minimize the context switching time.
 Provides concurrency within a process.
 Efficient communication.
 Allow utilization of multiprocessor architectures to

a greater scale and efficiency.

 Types of Thread
 User Level Threads
 Types of Thread (continue …)
 User Level Threads (Continue…)
 Managed by users.

 Kernel is not aware of the existence of threads.

 Multithreading is not compatible
 The thread library contains code for creating and

destroying threads.
 Doesn’t recognized by OS .
 Take less time for context switching.
 Can be created and managed quickly.
 Example: Java thread
 Types of Thread (Continue . . .)
 Advantages
 Simple and quick to create.

 Can run on any operating system

 They don’t need system calls to create threads.

 Switching between threads does not need kernel mode

privileges.
 If a thread in the kernel is blocked, it does not block all other

threads in the same process.

 Disadvantages
 Cannot benefit from multiprocessing.

 If a thread performs a blocking operation, the entire process is

halted.
 Types of Thread (Continue . . .)
 Kernel level threads (Continue . . .)
 Managed by OS.
 Multithreading is compatible
 Complicated implementation.
 Take more time to context switching
 Hardware support is needed.
 If one kernel thread performs a blocking operation

then another thread can continue execution.

 If a thread in the kernel is blocked, it does not block

all other threads in the same process.

 Multithreading
 Multithreading refers to the ability of an OS to support
multiple threads of execution within a single process.
 If a process has multiple thread of control, it can
perform more than one task at a time
 Each threads have their own program counter, stacks
and registers; But they share common code, data and
some OS data structures like files.
 Thread Models
 Mapping of user level threads to kernel level threads.
 The user threads maps to kernel threads in one of the
following strategies:
 One to one model
 Many to one model
 Many to Many models
 Thread Models (continue …)
 One to one model
 Maps a single user to a single kernel-level thread.
 Thread Models (continue …)
 Allows multiple threads to run in parallel on multiprocessors
 Drawbacks: The generation of every new user threads must

include creating a corresponding kernel threads causing an

overhead.
 Many to one
 Maps many user levels threads to one kernel thread.

 Thread management is handled by the thread library in user

space, which is very efficient.
 Facilitates an effective context-switching environment.

 Disadvantage:

‣ If a blocking system call is made, then the entire process

blocks.
Thread Models (continue …)
 Sinceonly one thread can access a kernel at a time, multiple
threads are unable to in parallel on multiprocessors

Fig : Many-to-one model

 Thread Models (continue …)
 Many to many model
 Multiplexes any number of user threads onto an equal or
smaller number of kernel threads.
 Combining the best features of the one-to-one and many-
to-one models.
 Users can create any number of threads.
 When a thread performs, a blocking system call, the
kernel can schedule another thread for execution.
 Processes can be split across multiple processors.
Thread Models (continue …)

Fig: Many to many model

 Inter process communication (IPC)
 Processes frequently need to communicate with other
processes.
 IPC is the way in which process communicate with
each other.
 The process concurrently executing in OS either
Independent or Cooperating process.
A. Independent process: The execution of one process
does not affect the execution of another process.
B. Cooperating process: The execution of one process
affects the execution of another process.
 IPC (continue…)
‣These processes need to be synchronized so that the
order of execution can be guaranteed.
 Reasons for cooperation
 Information sharing: for example several users

interested to access files concurrently.

 Computation speedup: to computational speed up

divide the task into several subtasks. All those subtasks

made to run simultaneously by assigning several
different processors.
 IPC (continue…)
 Modularity: Dividing the system in to different
modules and later modules put together to achieve the
single goal. Those modules should communicate or
cooperate to each other.
 Convenience: While running different tasks at the

same time, should be run smoothly without crash each

other.
 IPC (continue…)
 There are two fundamental models of IPC
A. Shared memory: The region of memory shared by
cooperating processes. Processes can exchange
information by writing and reading to and from the
shared region.
B. Message passing: Communication takes place by
means of exchange messages between the cooperating
processes.
‣ This is particularly useful for distributed environment,
where the communicating process reside different
computers connected by a network.
IPC (continue…)

Fig: a Shared memory Fig: b Message passing

 Race condition
 A race condition is a situation that may occur inside a
critical section.
 This happens when the result of multiple thread
execution in critical section deviate according to the
order in which the threads execute.
 Proper thread synchronization variables can prevent
race conditions.
 This leads to data inconsistency
 The figure in the next slide shows how race condition
can occur.
Race condition (Continue . . .)

Fig: Race condition

 Critical Section (CS)
 In multiprogramming, to avoid erroneous behavior the
shared resource need to be protected.
 The protected section is critical section.
 It can’t be executed more than one process at a time.
 the entry sections handles the entry into the critical section.
It acquires the resources needed for execution by the
process.
 The exit section handles the exit from the critical section. It
releases the resources and also informs the other processes
that critical section is free.
 The critical section problem needs a solution to
synchronize the different processes.
 Critical Section (Continue…)
 The solution to the critical section problem must satisfy the
following conditions:
 Mutual Exclusion: This implies that only one process can
be inside the critical section at any time. If any other
processes require the critical section, they must wait until it
is free.
 Progress : means that if a process is not using the critical
section, then it should not stop any other process from
accessing it. In other words, any process can enter a critical
section if it is free.
 Bounded Waiting: Means that each process must have a

limited waiting time. It should not wait endlessly to access

the critical section.