On This Page
This set of Multi-core processors Multiple Choice Questions & Answers (MCQs) focuses on Multi Core Processors Set 2
Q1 | Let S and Q be two semaphores initialized to 1, where P0 and P1 processes the following statements wait(S);wait(Q); ---; signal(S);signal(Q) and wait(Q); wait(S);---;signal(Q);signal(S); respectively. The above situation depicts a .
- . livelock
- . critical section
- . deadlock
- . mutual exclusion
Q2 | Which of the following conditions must be satisfied to solve the critical section problem?
- . mutual exclusion
- . progress
- . bounded waiting
- . all of the mentioned
Q3 | a process is allowed to enter its critical section .
- . after a process has made a request to enter its critical section and before the request is granted
- . when another process is in its critical section
- . before a process has made a request to enter its critical section
- . none of the mentioned
Q4 | processes to solve the critical section problem.
- . one
- . two
- . three
- . four
Q5 | What are Spinlocks?
- cpu cycles wasting locks over critical sections of programs
- locks that avoid time wastage in context switches
- locks that work better on multiprocessor systems
- all of the mentioned
Q6 | What is the main disadvantage of spinlocks?
- they are not sufficient for many process
- they require busy waiting
- they are unreliable sometimes
- they are too complex for programmers
Q7 | at any moment (the mutex being initialized to 1)?
- 1
- 2
- 3
- none of the mentioned
Q8 | Here, w1 and w2 have shared variables, which are initialized to false. Which one of the following statements is TRUE about the above construct?
- it does not ensure mutual exclusion
- it does not ensure bounded waiting
- it requires that processes enter the critical section in strict alternation
- it does not prevent deadlocks but ensures mutual exclusion
Q9 | The signal operation of the semaphore basically works on the basic system call.
- continue()
- wakeup()
- getup()
- start()
Q10 | (not necessarily immediately)?
- . #pragma omp section
- . #pragma omp parallel
- . none
- . #pragma omp master
Q11 | When compiling an OpenMP program with gcc, what flag must be included?
- . -fopenmp
- . #pragma omp parallel
- . –o hello
- . ./openmp
Q12 | program, i.e., each thread executes the same code.
- . parallel
- . section
- . single
- . master
Q13 | they would be in serial program.
- . nowait
- . ordered
- . collapse
- . for loops
Q14 | code of functions that are called (directly or indirectly) from within the parallel region.
- . lexical extent
- . static extent
- . dynamic extent
- . none of the above
Q15 | The specifies that the iterations of the for loop should be executed in parallel by multiple threads.
- . sections construct
- . for pragma
- . single construct
- . parallel for construct
Q16 | active in the parallel section region.
- . omp_get_num_procs ( )
- . omp_get_num_threads ( )
- . omp_get_thread_num ( )
- . omp_set_num_threads ( )
Q17 | The size of the initial chunksize .
- . total_no_of_iterations / max_threads
- . total_no_of_remaining_iterations / max_threads
- . total_no_of_iterations / no_threads
- . total_no_of_remaining_iterations / no_threads
Q18 | parallel architectures affect parallelization?
- . performance
- . latency
- . bandwidth
- . accuracy
Q19 | global_count += 5;
- . 4 instructions
- . 3 instructions
- . 5 instructions
- . 2 instructions
Q20 | generate log files of MPI calls.
- mpicxx
- mpilog
- mpitrace
- mpianim
Q21 | selectively screen messages.
- dest
- type
- address
- length
Q22 | them in this case and returning the result to a single process.
- mpi _ reduce
- mpi_ bcast
- mpi_ finalize
- mpi_ comm size
Q23 | The easiest way to create communicators with new groups is with .
- mpi_comm_rank
- mpi_comm_create
- mpi_comm_split
- mpi_comm_group
Q24 | is an object that holds information about the received message, including, for example, it’s actually count.
- buff
- count
- tag
- status
Q25 | The operation similarly computes an element-wise reduction of vectors, but this time leaves the result scattered among the processes.
- reduce-scatter
- reduce (to-one)
- allreduce
- none of the above