Distributed Operating Systems CS551

1
Distributed Operating SystemsCS551

• Colorado State University
• at Lockheed-Martin
• Lecture 2 -- Spring 2001

2
CS551 Lecture 2

• Topics
• Interconnection Networks
• Transparencies
• Real Time Systems (and networks)
• Parallel Systems (architecture software)
• Kernel of a Distributed Operating System
• Processes and Threads
• Interprocess Communication

3
Interconnection networks

• A distributed system a collection of autonomous
  computers linked by a computer network with
  distributed system software Coulouris,
  Dollimore Kindberg, Distributed Systems
  Concepts and Design, 2nd ed., Addison-Wesley
  (1994).
• Interconnection network designs
• bus, crossbar, hypercube, shuffle-exchange

4
Figure 1.8 An Interconnection Network in a
Multiprocessor. (Galli,p.20)
5
Interconnection network Bus

• Simple, static
• Shared by all attached processors
• No routing required
• All data transmissions occur on bus
• Scalability limited by bandwidth
• Examples
• Encore Multimax
• Sequent Balance and Sequent Symmetry

6
Figure 1.9 A Bus Used as an Interconnection
Network. (Galli, p.20)
7
Interconnection network Crossbar

• Permits dynamic configurations
• n memory modules x n processors
• Requires O(n2) switches (expensive)
• Many paths from one processor to another

8
Figure 1.10  A Crossbar Interconnection Network.
(Galli, p.22)
9
Interconnection network Hypercube

• Static
• Requires 2k switches for k processors
• Each switch is within a processor node
• Distance between nodes easy to compute

10
Figure 1.11 A Hypercube Inter-connection
Network of Order 3. (Galli.p.22)
11
Interconnection network Shuffle-Exchange

• Permits dynamic reconfiguration
• Switches have four modes of operation
• controlled by control bits in switch
• Simple routing
• address bits used as control bits
• Examples
• Benes, Banyan, Omega, Theta

12
Figure 1.12  A 2x2 Switch Box in a
Shuffle-Exchange Interconnection Network.
(Galli,p.23)
13
Figure 1.13  A Perfect Shuffle. (Galli,p.24)
14
Figure 1.14  A Multistage Shuffle-Exchange
Interconnection Network. (Galli,p.24)
15
Distributed Systems Characteristics

• Shared resources
• Openness
• Concurrency
• Scalability
• Fault tolerance
• Transparency

16
D.S. Characteristics Openness

• a characteristic that enables systems to be
  extended to meet new application requirements and
  user needs (CDK, p.12)
• achieved by specifying and documenting the key
  software interfaces of a system and making them
  available to software developers i.e. the
  interfaces are published (CDK, p.14)

17
D.S. Characteristics Transparency

• the concealment from the user and from the
  application programmer of the separation of
  components in a distributed system, so that the
  system is perceived as a whole rather than as a
  collection of independent components (CDK, p.20)
• Nine types
• access, location, concurrency, replication,
  failure, migration, name, performance, scaling

18
Network transparency

• Network transparency access transparency
  location transparency
• Access transparency enables local and remote
  information objects to be accessed using
  identical operations (CDK, p.20) Location
  transparency enables information objects to be
  accessed without knowledge of their location
  (CDK, p.20)

19
Concurrency transparency

• Concurrency and parallelism transparency
• enables several processes to operate
  concurrently using shared information objects
  without interference between them (CDK, p.20)
• permits efficient use of shared resources
• no interference between processes sharing
  resources

20
Replication transparency

• Replication transparency
• enables multiple instances of information
  objects to be used to increase reliability and
  performance without knowledge of the replicas by
  users of application programs (CDK, p.20)
• may be multiple copies of files in the system
• updates to such files can be a problem

21
Failure transparency

• Failure transparency
• enables the concealment of faults, allowing
  users and application programs to complete their
  tasks despite the failure of hardware or software
  components (CDK, p.20)
• if a site goes down, it is unapparent to other
  sites -- and work continues

22
Migration transparency

• Migration transparency
• allows the movement of information objects
  within a system without affecting the operation
  of users or application programs (CDK, p.21)
• both resources and processes can migrate without
  users knowing

23
Name transparency

• Name transparency
• system incorporates a global naming scheme
• objects (files, resources) are not tied to given
  nodes or sites by name
• assists migration, access, and location
  transparencies

24
Performance and Scaling transparencies

• Performance transparency
• allows the system to be reconfigured to improve
  performance as loads vary (CDK, p.21)
• Scaling transparency
• allows the system and applications to expand in
  scale without change to the system structure or
  the application algorithms (CDK, p.21)

25
Distributed Operating Systems

• A distributed operating system is
• a collection of software components that
  simplifies the task of programming and supports
  the widest possible range of applications (CDK,
  p.5)

26
Kernel of an operating system

• The privileged portion of the OS
• complete access to all resources
• controls
• process management
• process migration
• process scheduling
• address space

27
Types of OS kernels

• Monolithic kernel
• Unix, OS/360, VMS
• every node doesnt need entire kernel in DOS
• Layered kernel
• Windows NT/2000, Mach, Chorus (JavaOS)
• Collective kernel in DOS
• O.S. services are processes
• microkernel supports messages between such
  processes

28
Figure 2.1 Microkernel Design. (Galli,p.32)
29
Processes

• Process
• a program whose execution has started, but not
  terminated
• has state (ready, running, waiting)
• has a single address space
• may run serially or concurrently
• may interact with other processes via
• shared memory
• message passing
• is a single thread of control

30
Figure 2.4 Process States. (Galli,p.39)
31
Threads

• Thread
• is a lightweight process
• has state
• may share address space with other threads
• may run serially or concurrently
• interacts with other threads via
• shared address space
• message passing

32
Figure 2.2 A Multithreaded Process.
(Galli,p.34)
33
Issues with processes and threads

• Shared memory space
• no protection barrier to other processes/threads
• need to maintain integrity
• Synchronized access
• must enforce mutual exclusion
• code that accesses a shared resource must be a
  critical section

34
Multi-threaded systems

• Three models
• specialist -- all threads equal
• client/server -- server allots tasks to clients
• assembly line -- like a pipeline
• Supporting
• POSIX
• Java

35
Figure 2.3 Multithreaded Process Paradigms.
(Galli,p.35)
36
Why threads?

• Processes
• to create or destroy a process is expensive
• requires more memory space
• to restore or swap out a process is expensive
• requires memory map changes
• Threads
• can keep a pool of threads and reuse them
• memory space is shared and need not always be
  swapped

37
Process management

• Controls process (or thread) and its components
• Recall PCB (Process Control Block)
• process id
• process state
• process priority
• process privileges
• virtual memory address
• recorded statistics for account

38
Figure 2.4 Process States. (Galli,p.39)
39
Processes, continued

• Indivisible
• independent
• not divisible into smaller tasks
• Divisible
• may be broken up into smaller processes (tasks)
• subtasks may run on different nodes
• helps to balance load of distributed system
• Task Interaction Graphs (TIG)

40
Example of TIG

• TIG shows precedence of subtasks
• T1 -- Prepare main course
• T2 -- Prepare side dishes
• T3 -- Prepare dessert
• T4 -- Set table
• T5 -- Serve meal
• T6 -- Serve dessert

41
Figure 2.5 TIG Revealing Precedence
Relationships. (Galli,p.42)
42
Load Distribution

• Goal
• to utilize resources in an efficient manner
• Load balancing
• to balance the load equally among resources
• Load sharing
• to relieve overloaded resources
• Process migration
• to move a process to another processor

43
Load distribution algorithms

• Two parts
• Information-gathering
• status states (site is overloaded/underutilized?)
  
• Process selection
• which process/thread/task to migrate?
• expected overhead for migration?
• expected execution time?

44
Heterogeneous environments

• May include different data representations
• process migration may require data translation
• External data representation
• common data representation between sites

45
Figure 2.6 Data Translation without External
Data Representation. (Galli, p.44)
46
Figure 2.7 Data Translation with External Data
Representation. (Galli,p.44)
47
Process Scheduling

• Recall from Operating Systems course
• Problem 2.10 in textbook
• Identifying schedulable tasks
• polled loop
• interrupts
• gang scheduling

48
Figure 2.8 Polled-Loop Example. (Galli,p.47)
49
Figure 2.9 Smart Concurrent Scheduling.
(Galli,p.48)
50
Figure 2.10  Blind Concurrent Scheduling.
(Galli,p.48)
51
Remote procedure calls (RPC)
52
Figure 3.9  Remote Procedure Call Stubs.
(Galli,p.73)
53
Figure 3.10  Establishing Communication for
RPC. (Galli,p.74)