Replication and caching for performance are widely applied as __________ techniques.

• A) scaling
• B) deadlock
• C) context switching
• D) concurrency

________ is NVIDIA’s general purpose parallel computing architecture designed for calculation-intensive computation on GPU hardware.

• A) CAD
• B) SoC
• C) CUDA
• D) DEC

Which of the following has emerged as the standard threads API, supported by most vendors?

• A) POSIX threads
• B) Java threads
• C) Solaris threads
• D) NT threads

__________ is a non-blocking network allowing multiple input output connection pattern to be achieved simultaneously.

• A) Cross Bar
• B) Mesh
• C) Tree
• D) Bus

The finer decomposition of tasks always results in small time.

• A) False
• B) True
• C)
• D)

The _________ of a network with n nodes is the length of the maximum shortest path between any two nodes in the network.

• A) latency
• B) bandwidth
• C) diameter
• D) bisection

Which of the following is true regarding thread accept?

• A) Can be scheduled to run on a CPU as an independent unit
• B) Lightweight process
• C) Executes within the address space of a process
• D) Can not run on CPU as independent unit

__________ is the process by which data residing on a physical/virtual servers or cloud instance is continuously copied to a secondary server(s) or cloud instance.

• A) Data replication
• B) Deadlock
• C) Scalability
• D) Concurrency

There are _________ commonly used decomposition techniques.

• A) Three
• B) Five
• C) Seven
• D) Four

__________ is a specification for the developers and users of message passing libraries.

• A) Graphics Processing Interfacet (GPI)
• B) Message Passing Unit (MPU)
• C) Graphics Processing Unit (GPU)
• D) Message Passing Interface (MPI)

_________ is process to forward packets coming in from one port to a port leading towards the destination.

• A) Latency
• B) Context switching
• C) Temporal locality
• D) Switching

The problem of finding the minimum in a list is an example of decomposition of data based on _______________.

• A) both input and output data
• B) output data
• C) input data
• D) intermediate data

________ of Stanford University considers power and massive parallelism as the major benefits of GPUs over CPUs for the future.

• A) Bill Dally
• B) Geoffrey Hinton
• C) John McCarthy
• D) Larry Page

_________ consist of point-to-point communication links among processing nodes and are also referred to as direct networks.

• A) Bus networks
• B) Dynamic networks
• C) Indirect networks
• D) Static networks

____________ is a special class of distributed data-store which is characterized by the lack of simultaneous updates. Here, the emphasis is more on maintaining a consistent view of things for the individual client process that is currently operating on the data-store.

• A) Sequential model
• B) Data consistency model
• C) Client centric consistency model
• D) Incremental model

Number of bits transmitted in a unit of time is known as____________.

• A) bisection
• B) diameter
• C) bandwidth
• D) latency

In _________ communication, the sender waits for a response from the receiver before proceeding with further execution. This ensures that the sender and receiver are in sync with each other.

• A) synchronous
• B) buffered
• C) unbuffered
• D) asynchronous

Which of the following option is not the communication mode that can be used for message passing programming?

• A) Buffered/unbuffered
• B) Verbal/non-verbal
• C) Blocking/non-blocking
• D) Synchronous/asynchronous

In __________ control strategy, one central control unit is used to control the operations of the components of the system.

• A) distributed
• B) supervised
• C) centralized
• D) decentralized

___________ is a method of communication in distributed systems where processes or objects exchange messages with each other to share information or coordinate activities.

• A) Deadlock
• B) Concurrency
• C) Caching
• D) Message passing