Alan Kaminsky • Department of Computer Science • Rochester Institute of Technology • 4486 + 2220 = 6706
Home Page
Operating Systems I • 4003-440-02 • Winter Quarter 2012
Course Page

4003-440-02 Operating Systems I
Module 2. Process Management -- Lecture Notes

Prof. Alan Kaminsky -- Winter Quarter 2012
Rochester Institute of Technology -- Department of Computer Science

Processes and Threads

  • Thread
    • "Thread of execution" -- why is it called a "thread"?
    • Resources associated with a thread
      • CPU registers (PC, SP, etc.)
      • Code (text)
      • Data
      • Stack
      • Heap
      • OS resources (e.g. files)
         
  • Process
    • A group of threads sharing certain resources
    • Resources shared by all threads in a process
      • Code (text)
      • Global data
      • Heap
      • OS resources (e.g. files)
    • Resources private to each thread in a process
      • CPU registers (PC, SP, etc.)
      • Thread local data
      • Stack
         
  • Threads (within one process) = Same address space
     
  • Processes = Different address spaces

Process and Thread Scheduling

  • Mapping of user threads to kernel threads
    • One-to-one model -- used by most modern OSes
    • Many-to-one model -- historical relict (e.g. Green Threads)
    • Many-to-many model
       
  • Thread state diagram
    • Ready
    • Running
    • Waiting
       
  • Thread scheduling queues
    • Ready queue
    • Wait queues
      • I/O completion
      • Timer
      • Thread coordination
      • Thread termination
      • Child process termination
      • Etc.
         
  • Thread scheduling on a single processor
    • Task control block (TCB)
    • Context switching
    • Multiple threads pattern of execution
       
  • Thread scheduling on a multiprocessor
    • Multiple threads pattern of execution

Thread Coordination

  • Threads (within a process) can communicate "for free" -- shared memory (same address space)
     
  • Thread coordination
    • Exclusive access to shared data
    • Read or write access to shared data
    • Waiting for another thread to perform some action

Interprocess Communication

  • Different processes cannot communicate "for free" -- different address spaces
     
  • IPC via shared memory
    • Map the same region of physical memory into the address spaces of multiple processes
    • Doesn't make much sense nowadays -- just use threads
       
  • IPC via message passing
    • One process explicitly sends a message, another process explicitly receives the message
    • Nowadays it usually makes more sense to use . . .
       
  • IPC via networking protocols
    • TCP sockets, UDP datagrams
    • Works between processes in the same computer (localhost) and between processes in different computers
    • Same software works on all architectures from single processor to distributed system
    • Platform independent software (Java)
       
  • IPC via remote procedure call (RPC)
     
  • IPC via remote method invocation (RMI)
    • CORBA
    • Java RMI
    • Microsoft .NET remoting

Operating Systems I • 4003-440-02 • Winter Quarter 2012
Course Page
Alan Kaminsky • Department of Computer Science • Rochester Institute of Technology • 4486 + 2220 = 6706
Home Page
Copyright © 2008 Alan Kaminsky. All rights reserved. Last updated 27-Aug-2008. Please send comments to ark­@­cs.rit.edu.