Q.1
These applications typically have multiple executable object files (programs). While the application is being run in parallel, each task can be executing the same or different program as other tasks. All tasks may use different data
  • Single Program Multiple Data (SPMD)
  • Multiple Program Multiple Data (MPMD)
  • Von Neumann Architecture
  • None of these
Q.2
In the threads model of parallel programming
  • A single process can have multiple, concurrent execution paths
  • A single process can have single, concurrent execution paths.
  • A multiple process can have single concurrent execution paths.
  • None of these
Q.3
It distinguishes multi-processor computer architectures according to how they can be classified along the two independent dimensions of Instruction and Dat(A) Each of these dimensions can have only one of two possible states: Single or Multiple.
  • Single Program Multiple Data (SPMD)
  • Flynn’s taxonomy
  • Von Neumann Architecture
  • None of these
Q.4
Non-Uniform Memory Access (NUMA) is
  • Here all processors have equal access and access times to memory
  • Here if one processor updates a location in shared memory, all the other processors know about the updat
  • Here one SMP can directly access memory of another SMP and not all processors have equal access time to all memories
  • None of these
Q.5
Cache Coherent UMA (CC-UMA) is
  • Here all processors have equal access and access times to memory
  • Here if one processor updates a location in shared memory, all the other processors know about the updat
  • Here one SMP can directly access memory of another SMP and not all processors have equal access time to all memories
  • None of these
Q.6
Coarse-grain Parallelism
  • In parallel computing, it is a qualitative measure of the ratio of computation to communication
  • Here relatively small amounts of computational work are done between communication events
  • Relatively large amounts of computa- tional work are done between communication / synchronization events
  • None of these
Q.7
Granularity is
  • In parallel computing, it is a qualitative measure of the ratio of computation to communication
  • Here relatively small amounts of computational work are done between communication events
  • Relatively large amounts of computa- tional work are done between communication / synchronization events
  • None of these
Q.8
Asynchronous communications
  • It involves data sharing between more than two tasks, which are often specified as being members in a common group, or collectiv
  • It involves two tasks with one task acting as the sender/producer of data, and the other acting as the receiver/consumer.
  • It allows tasks to transfer data independently from one another.
  • None of these
Q.9
Uniform Memory Access (UMA) referred to
  • Here all processors have equal access and access times to memory
  • Here if one processor updates a location in shared memory, all the other processors know about the updat
  • Here one SMP can directly access memory of another SMP and not all processors have equal access time to all memories
  • None of these
Q.10
Point-to-point communication referred to
  • It involves data sharing between more than two tasks, which are often specified as being members in a common group, or collectiv
  • It involves two tasks with one task acting as the sender/producer of data, and the other acting as the receiver/consumer.*
  • It allows tasks to transfer data independently from one another.
  • None of these
Q.11
Collective communication
  • It involves data sharing between more than two tasks, which are often specified as being members in a common group, or collectiv
  • It involves two tasks with one task acting as the sender/producer of data, and the other acting as the receiver/consumer.
  • It allows tasks to transfer data independently from one another.
  • None of these
Q.12
Synchronous communications
  • It require some type of “handshaking” between tasks that are sharing dat(A) This can be explicitly structured in code by the programmer, or it may happen at a lower level unknown to the pro- grammer.
  • It involves data sharing between more than two tasks, which are often specified as being members in a common group, or collectiv
  • It involves two tasks with one task acting as the sender/producer of data, and the other acting as the receiver/consumer.
  • It allows tasks to transfer data independently from one another.
Q.13
Functional Decomposition:
  • Partitioning in that the data associated with a problem is decompose(D) Each parallel task then works on a portion of the dat(A)
  • Partitioning in that, the focus is on the computation that is to be performed rather than on the data manipulated by the computation. The problem is decomposed according to the work that must be don Each task then performs a portion of the overall work.
  • It is the time it takes to send a minimal (0 byte) message from point A to point (B)
  • None of these
Q.14
Domain Decomposition
  • Partitioning in that the data associated with a problem is decompose(D) Each parallel task then works on a portion of the dat(A)
  • Partitioning in that, the focus is on the computation that is to be performed rather than on the data manipulated by the computation. The problem is decomposed according to the work that must be don Each task then performs a portion of the overall work.
  • It is the time it takes to send a minimal (0 byte) message from point A to point (B)
  • None of these
Q.15
Latency is
  • Partitioning in that the data associated with a problem is decompose(D) Each parallel task then works on a portion of the dat(A)
  • Partitioning in that, the focus is on the computation that is to be performed rather than on the data manipulated by the computation. The problem is decomposed according to the work that must be don Each task then performs a portion of the overall work.
  • It is the time it takes to send a minimal (0 byte) message from one point to other point
  • None of these
Q.16
In designing a parallel program, one has to break the problem into discreet chunks of work that can be distributed to multiple tasks. This is known as
  • Decomposition
  • Partitioning
  • Compounding
  • Both A and B
Q.17
In shared Memory
  • Changes in a memory location effected by one processor do not affect all other processors.
  • Changes in a memory location effected by one processor are visible to all other processors
  • Changes in a memory location effected by one processor are randomly visible to all other processors.
  • None of these
Q.18
Fine-grain Parallelism is
  • In parallel computing, it is a qualitative measure of the ratio of computation to communication
  • Here relatively small amounts of computational work are done between communication events
  • Relatively large amounts of computational work are done between communication / synchroni- zation events
  • None of these
Q.19
Massively Parallel
  • Observed speedup of a code which has been parallelized, defined as: wall-clock time of serial execution and wall-clock time of parallel execution
  • The amount of time required to coordinate parallel tasks. It includes factors such as: Task start-up time, Synchronizations, Data communications.
  • Refers to the hardware that comprises a given parallel system - having many processors
  • None of these
Q.20
Parallel Overhead is
  • Observed speedup of a code which has been parallelized, defined as: wall-clock time of serial execution and wall-clock time of parallel execution
  • The amount of time required to coordi- nate parallel tasks. It includes factors such as: Task start-up time, Synchro- nizations, Data communications.
  • Refers to the hardware that comprises a given parallel system - having many processors
  • None of these
0 h : 0 m : 1 s