Chapter 4: Processes 9/16/03

1
Chapter 4 Processes9/16/03

• Process Concept
• Process Scheduling
• Operations on Processes
• Cooperating Processes
• Interprocess Communication
• Communication in Client-Server Systems
• NOTE Instructor annotations in BLUE

2
Process Concept

• An operating system executes a variety of
  programs
• Batch system jobs
• Time-shared systems user programs or tasks
• Textbook uses the terms job and process almost
  interchangeably.
• Process a program in some state of execution
  process execution must progress in sequential
  fashion.
• A process includes
• program counter
• stack
• data section
• Text section (code)

3
Process State

• As a process executes, it changes state
• new The process is being created.
• running Instructions are being executed.
• waiting The process is waiting for some event
  to occur.
• ready The process is waiting to be assigned to
  a processor not waiting for anything or
  resource.
• terminated The process has finished execution.

4
Diagram of Process State
5
Queuing representation of Process States
A queue for each event is faster to search gt
From Stallings, Operating Systems, 4th ed. fig
3.8b, p. 121
6
The Suspended State

• Suspended State When a process which has been
  completely removed from memory and now is
  residing on the disk as a process image which can
  be reloaded into memory swapping.
• Non-virtual memory systems This state is always
  needed in a system which does not employ virtual
  memory uses process swapping.
• Process is either fully in memory or fully
  swapped out to disk.
• Needed if most processes in memory are blocked
  (say for I/O) and.
• By swapping out to the suspended state, memory is
  freed for a new or runnable processes.
• CPU utilization maximized.
• Virtual memory systems process is only partially
  in memory.
• May still be a need for swapping and the
  suspended state.
• If too many processes are resident in memory
  (even partially), performance is impacted
  thrashing low CPU utilization
• Thus swapping out a process to the disk(Suspended
  state) will improve performance.

7
Suspended States
Most general state diagram
From Stallings, Operating Systems, 4th ed. fig
3.8b, p. 121
8
Transitions involving suspended states

• Blocked -gt Blocked/Suspend all processes
  waiting, no ready processes, blocked process is
  swapped out to make room for unblocked processes
• Blocked/Suspend -gtReady/Suspend event waited
  for occurs, but still remains in suspended state
  maybe memory space unavailable.
• Ready/Suspend -gt Ready all processes waiting,
  swap in a ready/ suspended processes.
• Ready-gt Ready/Suspend Rare transition, if
  running process is extremely large to free up
  memory, or due to priorities.
• New-gt Ready/Suspend Maintain a pool of
  non-blocked processes on disk overhead of
  building process done in advance
• Blocked/Suspend -gtBlocked System anticipates
  that waited event will occur soon move blocked
  process into memory.
• Running -gt Ready/Suspend Preempt a running
  process due to a high priority blocked process
  becoming unblocked

9
Process Control Block (PCB)

• Information associated with each process -
  uniquely
• identifies and characterizes a particular
  process.
• Some information in the PCB
• Process state running, waiting, ready, etc.
• Program counter
• CPU register values
• CPU scheduling information - used by short term
  scheduler - see later
• Memory-management information base, limit
  registers, page table info, etc.
• Accounting information
• I/O status file information devices allocated
  to proc., open files, etc.

10
Process Control Block (PCB)
11
CPU Switch From Process to Process
(context switch)
Process state saved in PCB
12
Process Scheduling Queues

• Job queue set of all processes in the system.
• Ready queue set of all processes residing in
  main memory, ready and waiting to execute.
• Device queues set of processes waiting for an
  I/O device.
• Process migration between the various queues.
  Elements entered in these queues are PCBs
  linked list

13
Ready Queue And Various I/O Device Queues
14
Representation of Process Scheduling
15
Schedulers

• Long-term scheduler (or job scheduler) selects
  which processes should be brought into the ready
  queue.Schedule new jobs to become a process
  (from disk)
• Short-term scheduler (or CPU scheduler) selects
  which process should be executed next and
  allocates CPU.
• Medium term scheduler - remove processes,
  especially suspended, from memory and move to
  disk. Later bring process back into memory
• Swapping - controls degree of Multiprogramming,
  and memory utilization

16
Addition of Medium Term Scheduling
17
Schedulers (Cont.)

• Short-term scheduler is invoked very frequently
  (milliseconds) ? (must be fast) a potential
  performance bottleneck).
• Long-term scheduler is invoked very infrequently
  (seconds, minutes) ? (may be slow).
• The long-term medium term schedulers controls
  the degree of multiprogramming how many
  programs to have in memory at one time.
• Has implications on controlling thrashing in a
  Virtual memory system
• Processes can be described as either
• I/O-bound process spends more time doing I/O
  than computations, many short CPU bursts.
• CPU-bound process spends more time doing
  computations few very long CPU bursts.

18
Context Switch

• When CPU switches to another process, the system
  must save the state of the old process and load
  the saved state for the new process.
• Context-switch time is overhead the system does
  no useful work while switching an area needing
  optimization.
• Time dependent on hardware support.

19
Process Creation

• Parent process create children processes, which,
  in turn create other processes, forming a tree of
  processes.Example UNIX fork/spawning mechanism.
• Resource sharing
• Parent and children share all resources.
• Children share subset of parents resources.
• Parent and child share no resources.
• Execution
• Parent and children execute concurrently.
• Parent waits until children terminate - wait is a
  system call.

20
Process Creation (Cont.)

• Address space
• Child duplicate of parent.
• Child has a program loaded into it.
• UNIX examples
• fork system call creates new process
• exec system call used after a fork to replace the
  process memory space with a new program.
• fork and execve are both system calls
• Except for 3 special processes that are part of
  the kernel scheduler, init, and pagedaemon (PIDs
  0, 1, 2), all processes are create from a fork
  call.
• All other processes have fork ancestries which
  start with the init system process
• A typical user program run from the command line
  would be forked from the shell process.
• If the parent quits before waiting for a child to
  quit, the orphan child is adopted by init and
  becomes its child.
• If child quits before the parent waits for it
  to quit, the child then becomes a zombie the
  zombie is required for the parent to successfully
  do a wait for the child (and get certain status).
  Zombie goes away when wait occurs.
• Since init never ends, it avoids clogging up the
  system with zombies by automatically calling wait
  when a child quits.

21
Processes Tree on a UNIX System
swapper is the scheduler
the family tree
22
Process Termination

• Process executes last statement and asks the
  operating system to decide it (exit).
• Output data from child to parent (via wait).
• Process resources are deallocated by operating
  system.
• Parent may terminate execution of children
  processes (abort).
• Child has exceeded allocated resources.
• Task assigned to child is no longer required.
• Parent is exiting.
• Operating system does not allow child to continue
  if its parent terminates in UNIX it child is
  adopted by init.
• Cascading termination.
• In UNIX terminated process may become a Zombie
  before going away - these may haunt the system
  for a while! (see earlier slide).

23
Cooperating Processes

• Independent process cannot affect or be affected
  by the execution of another process.
• Cooperating process can affect or be affected by
  the execution of another process
  synchronization required
• Advantages of process cooperation
• Information sharing
• Computation speed-up - only when distributed
  across multiple processors
• Modularity - same rationale as used for
  strucuring a program along function/subroutine or
  OO class lines.
• Convenience

24
Producer-Consumer Problem

• Example of Cooperating Processes
• Paradigm for cooperating processes, producer
  process produces information that is consumed by
  a consumer process.
• unbounded-buffer places no practical limit on the
  size of the buffer.
• bounded-buffer assumes that there is a fixed
  buffer size.
• In both cases the buffer is shared memory
  shared by both processes

25
Bounded-Buffer Shared-Memory Solution

• Shared data
• define BUFFER_SIZE 10
• typedef struct
• . . .
• item
• item bufferBUFFER_SIZE
• int in 0
• int out 0
• int counter 0

26
Bounded-Buffer Producer Process

• Producer process
• item nextProduced
• while (1)
• while (counter BUFFER_SIZE)
• / do nothing /
• bufferin nextProduced
• in (in 1) BUFFER_SIZE
• counter

What if Consumer does not complete the
incrementing of shared variable counter? It
could take 3 instruction to increment and a task
switch could occur all three instructions are
complete?
27
Bounded-Buffer Consumer Process

• Consumer process
• item nextConsumed
• while (1)
• while (counter 0)
• / do nothing /
• nextConsumed bufferout
• out (out 1) BUFFER_SIZE
• counter--

What if Producer does not complete the
incrementing of shared variable counter? It
could take 3 instruction to increment and a task
switch could occur all three instructions are
complete?
28
Interprocess Communication (IPC)

• Mechanism for processes to communicate and to
  synchronize their actions - handshaking, null
  message scheme (see later).
• Message system processes communicate with each
  other without resorting to shared variables or
  common (shared) memory/buffer.
• IPC facility provides two operations(p. 111)
• send(message) message size fixed or variable
• receive(message)
• If P and Q wish to communicate, they need to
• establish a communication link between
  themmsgqid msgget() in UNIX project
• exchange messages via send/receive
• Implementation of communication link
• physical (e.g., shared memory, hardware bus,
  network) Use of shared memory simulates message
  passing - retains interface- covered in chapter
  15
• logical (e.g., logical properties or
  interface/API as seen by application) - only
  logical implementation covered here.

29
Implementation Questions

• How are links established? - explicitly vs.
  implicitly.
• Can a link be associated with more than two
  processes? - ex multiple processes communicating
  via a network.
• How many links can there be between every pair of
  communicating processes? Ex two processes may
  communicate via two message queues.
• What is the capacity of a link? (0, finite,
  infinite)
• Is the size of a message that the link can
  accommodate fixed or variable?
• Is a link unidirectional or bi-directional? - ex
  UNIX pipes can be either
• Blocking send/receive vs nonblocking send/receive
  - has synchronization implications.

30
Direct Communication

• Processes must name each other explicitly
• send (P, message) send a message to process P
• receive(Q, message) receive a message from
  process QName the other process in the call.
  P, Q are pids. In send, message can be a
  pointer or the message itself, in receive,
  message usually a pointer to where to put the
  received message.
• Properties of communication link
• Links are established automatically - only need
  to know the others identity.
• A link is associated with exactly one pair of
  communicating processes P1 ltL1gt P2 ltL2gtP3
• Between each pair there exists exactly one link.
• The link may be unidirectional, but is usually
  bi-directional. also asymmetric, receive from
  anyone, id of sender passed to receiver
  receive(id, message

31
Indirect Communication

• Messages are directed and received from mailboxes
  (also referred to as ports).
• Each mailbox has a unique message queue id.
• Processes can communicate only if they share a
  mailbox.
• Properties of communication link
• Link established only if processes share a common
  mailbox
• A link may be associated with many processes.
• Each pair of processes may share several
  communication links - P1 and P2 can communicate
  with each other via multiple message queues.
• Link may be unidirectional or bi-directional.

32
Indirect Communication - continued

• Operations
• create a new mailbox- msgget in UNIX
• send and receive messages through mailbox
• destroy a mailbox- msgctl() return to UNIX
• Primitives are defined as
• send(A, message) send a message to mailbox A
• receive(A, message) receive a message from
  mailbox AIn UNIX, mailbox A is the message Queue
  id returned by the msgget call

33
Indirect Communication - continued

• Mailbox sharing
• P1, P2, and P3 share mailbox A.
• P1, sends P2 and P3 receive.
• Who gets the message? - may be a race condition
• Solutions this is your problem!
• Allow a link to be associated with at most two
  processes.
• Allow only one process at a time to execute a
  receive operation.
• Allow the system to select arbitrarily the
  receiver. Sender is notified who the receiver
  was.
• Dont do anything - maybe it does not matter.
  The hungry dog approach.
• in UNIX can send a message class id receiver
  only receives messages in a certain class

34
Synchronization - sect 4.5.3

• Message passing may be either blocking or
  non-blocking.
• Blocking is considered synchronous
• Non-blocking is considered asynchronous
• send and receive primitives may be either
  blocking or non-blocking.

35
Buffering

• Queue of messages attached to the link
  implemented in one of three ways.
• 1. Zero capacity 0 messagesSender must wait
  for receiver (rendezvous).
• 2. Bounded capacity finite length of n
  messagesSender must wait if link full.
• 3. Unbounded capacity infinite length Sender
  never waits.

See notes by instructor Message Passing
Concepts and Synchronization for additional and
supplementary material on message passing. Read
pp. 1-4 for now.
36
Client-Server Communication

• EX A user (client) sends request for some
  information from a server which may also require
  some computation by the serverSome approaches
• Sockets
• Remote Procedure Calls
• Remote Method Invocation (Java)

37
Sockets

• A socket is defined as an endpoint for
  communication.
• Identified as a concatenation of IP address and
  port
• identifies a host and a port on that host.
• The socket 161.25.19.81625 refers to port 1625
  on host 161.25.19.8
• Communication consists between a pair of sockets.

38
Socket Communication
Client assigned port by the local host computer (
? 1024)
Server port is Well Known (lt 1024) Client port
is assigned a port the host it is running on
39
Remote Procedure Calls

• Remote procedure call (RPC) abstracts procedure
  calls between processes on networked systems.
• Message-based communication used to provide this
  facility.
• In contrast to general IPC message passing, RPC
  messages are highly structured and formalized.
• They are received by an RPC daemon, listening to
  a port and contains the identifier of the
  function to be called, and the parms to be
  passed.
• Stubs client-side proxy for the actual
  procedure on the server.
• Hides communication details and makes it look
  like an ordinary local procedural call.
• The client-side stub locates the server and
  marshalls the parameters (creates a network
  packet for parms).
• The server-side stub receives this message,
  unpacks the marshalled parameters, and performs
  the procedure on the server - returned values
  passed back to client using same technique.

40
Execution of RPC
Matchmaker gets server port to allow client
port to bind procedure name to a remote server
port. Client receives the server port number for
the given preocedure.
41
Remote Method Invocation

• Remote Method Invocation (RMI) is a Java
  mechanism similar to RPCs.
• RMI allows a Java program on one machine to
  invoke a method on a remote object.

42
Marshalling Parameters
Skeleton receives parcel, un-marshals it, invokes
the desired method, marshals results, and returns
parcel to client.
Stub is proxy for remote object. Creates a parcel
consisting of method name marshaled parms.