Ch 4

Question 1

Data parallelism

Answer 1

distributes subsets of the same data across multiple cores, same operation on each

Question 2

Task parallelism

Answer 2

distributing threads across cores, each thread performing unique operation

Question 3

User Threads

Answer 3

management done by user-level threads library i.e. POSIX Pthreads, Windows threads, Java threads

Question 4

Kernel Threads

Answer 4

virtually all general purpose operating systems, including Windows, Solaris, Linux/UNIX, Mac OS X

Question 5

Kernel Threads

Answer 5

virtually all general purpose operating systems, including Windows, Solaris, Linux/UNIX, Mac OS X

Question 6

Multithreading models

Answer 6

Many-to-One - Multiple threads may not run in parallel on muticore system because only one may be in kernel at a time

One-to-One - More concurrency than many-to-one. Number of threads per process sometimes restricted due to overhead

Many-to-Many - Allows the operating system to create a sufficient number of kernel threads

Question 7

Signals

Answer 7

Signals are used in UNIX systems to notify a process that a particular event has occurred.

Question 8

target thread

Answer 8

If thread has cancellation disabled, cancellation remains pending until thread enables it Asynchronous cancellation terminates the target thread immediately Deferred cancellation allows the target thread to periodically check if it should be cancelled

Question 9

TLS

Answer 9

Thread-local storage (TLS) allows each thread to have its own copy of data

Question 10

What data gets duplicated in threads?

Answer 10

Question 11

Benefits of multithreading?

Answer 11

• Responsiveness - useful for user interfaces
• Resource sharing - it allows an application to have several different threads of activity within the same address space.
• Economy - because they share recources
• Scalibility - multiprocessor architecture,

Question 12

Amdahl’s Law

Answer 12

Question 13

Challenges of multithreading?

Answer 13

Identifying tasks

balance

data splitting

Data dependency

Testing and Debugging

Question 14

thread pools

Answer 14

The general idea behind a thread pool is to create a number of threads at process startup and place them into a pool, to limit the amount of threads that can be in use (a multithread problem).

Servicing a request with an existing thread is faster.

Question 15

block

Answer 15

A block is simply a self-contained unit of work recognized by a compiler. ( i.e printf() )

Question 16

Signal handler

Answer 16

processes signals generated by a particular event, delivered to a process, handled

Question 17

thread cancellation

Answer 17

Asynchronous cancellation terminates the target thread immediately

asynchronous cancellation is problematic, because a thread may be terminated without letting the OS reclaim all of its recources

Deferred cancellation allows the target thread to periodically check if it should be canceled

deferred cancelation fixes the issue mentioned above.