Program

1
Program

• Whats a program?
• Is a static entity. Only exist as an idea, on
  paper, or as a file.

2
Processes

• Process An abstraction of a running program.
• Multiprogramming allows for the processor to
  switch between several different processes in
  memory.
• Processor may only run each program a fraction of
  a second

3
Pseudoparallelism

• Processor switches between each program so fast
  it gives the appearance of programs running in
  parallel.
• This is in contrast to true parallelism which
  requires a system with more than 1 CPU.

4
Thinking in Parallel

• Very hard for humans to do
• We have a conceptual model called the sequential
  process or just process.
• Associated with a process are its code, input,
  output and state.

5
Consider 3 processes

• How do we maintain where each process is in its
  execution as each one is switched out of running
  on the CPU?
• Each process has a logical PC that is saved with
  the process
• So, when a particular process is going to execute
  on the CPU what happens with the logical PC?
• Loaded into the Physical PC

6
Four Events that can trigger process creation

• System Initialization
• Execution of a process creation system call by a
  running process
• A user request to create a new process.
• Initiation of a batch process.

7
Types of Processes

• Foreground
• Processes that interact with the user and perform
  work for them
• Background
• Not associated with a user, but perform some type
  of specific function for the system (accepting
  incoming email)

8
After a process has performed its job

• It is terminated by the following methods
• Normal exit (voluntary)
• Error exit (voluntary Usually User Causes)
• Fatal error (involuntary Usually Process
  Causes)
• Killed by another process (involuntary)

9
Process States

• Running A process that is executing on a CPU.
  If the system has n CPUs, at most n processes may
  be in the running state.
• Ready A process that is not allocated to the
  CPU but is ready to run. A ready process could
  execute if allocated to a CPU.

10
Process States (cont.)

• Blocked A process that is waiting for some
  event to occur before it can continue executing.
  This event is often the completion of an I/O
  operation.
• Do not confuse Blocked and Ready Processes

11
Process State (cont)

• Terminated process that has halted its
  execution but a record of the process is still
  maintained by the OS. This record may be useful
  to other processes.

12
Process State Diagram
Terminated
Exit
Running
Scheduler decides another process should run
Process blocks waiting for some external event
Scheduler decides a process should run
Blocked
Ready
Once a process is created It is placed in the
ready state
Process is unblocked once external event occurs
New
13
So, how do we keep track of information about a
process?

• Process Control Block (PCB) Just an array of
  structures with 1 entry per process
• A PCB entry contains the PC, SP, memory
  allocation, status of its open files, scheduling
  info, anything to restore the process when
  switching states
• Nucleus Plus uses a PCBs but calls them another
  name
• Page 80 in the book has a more complete list of
  fields often found in a PCB

14
Two ways of looking at processes

• Resource Ownership
• Program (code), data, and other attributes
  defined in the PCB
• May be allocated other resources files, I/O
  devices, main memory
• Scheduling/execution
• Path of execution through a program
• Also a state (ready, running, blocked, etc)

15
Two ways of looking at processes

• We can separate these two concepts and many
  operating systems do separate these concepts
• Unit of dispatching/scheduling is called a thread
• Unit of resource ownership is referred to as a
  process

16
Different OS have different levels of thread
support
Instruction Trace
1 process, 1 thread - MSDOS
1 process, Multiple threads, Java RTE
Multiple process, 1 thread - UNIX
Multiple process, Multiple threads, Windows,
Solaris, OS/2
17
Processes vs. Threads

• Each process has its own address space
  containing program text and data along with other
  resources.
• A process also has a thread of execution, so
  threads are a part of a particular process.
• Each thread has its own PC, registers, stack,
  and state just like a process

18
Processes vs. Threads (cont.)

• But threads allow us to use the same
• Address space
• Global Variables
• Open Files
• Obviously we do not want to create separate
  processes (more overhead) if these resources can
  be shared.

19
Benefits of using Threads

• Ability of multithreaded entities to share
  address space and data.
• Easier to create and destroy in comparison to
  processes which have resources.
• Improved Performance when substantial computing
  power is available and substantial I/O
• Useful for multiprocessor systems

20
Thread Implementation User Level

• User-level threads perform threading operations
  in user space
• Threads are created by runtime libraries that
  cannot execute privileged instructions or access
  kernel primitives directly

21
Thread Implementation User Level

• User-level thread implementation
• Many-to-one thread mappings
• Operating system maps all threads in a
  multithreaded process to single execution context
• Advantages
• User-level libraries can schedule its threads to
  optimize performance
• Synchronization performed outside kernel, avoids
  context switches
• More portable Do not need OS support for
  threads
• Disadvantage
• Kernel views a multithreaded process as a single
  thread of control
• Can lead to suboptimal performance if a thread
  issues I/O
• Cannot be scheduled on multiple processors at
  once

22
Thread Implementation Kernel Level

• Kernel-level threads attempt to address the
  limitations of user-level threads by mapping each
  thread to its own execution context
• Kernel-level threads provide a one-to-one thread
  mapping
• Advantages Increased scalability, interactivity,
  and throughput
• Disadvantages Overhead due to context switching
  and reduced portability due to OS-specific APIs
• Kernel-level threads are not always the optimal
  solution for multithreaded applications