32 Questions
What is the main difference between static networks and dynamic networks?
Static networks consist of point-to-point communication links between processors, whereas dynamic networks are built using switches and communication links, and the communication links are connected to one another dynamically using the switches.
What is network latency, and how does it differ from communication latency?
Network latency is the time to transfer a message through the network, whereas communication latency is the transmission time plus overhead.
What is the significance of bisection width in evaluating a network?
Bisection width is the minimum number of links that must be cut to partition the network into two equal or almost equal halves, and it indicates potential communication bottlenecks.
What is the main distinction between shared-address-space machines and message-passing platforms?
Shared-address-space machines provide a shared data space, whereas message-passing platforms support messaging between parallel tasks.
What is the primary advantage of shared-address-space architectures?
Communication is implicitly specified since some (or all) of the memory is accessible to all the processors.
What is the role of switches in dynamic networks?
Switches are used to connect communication links dynamically in dynamic networks.
What is bandwidth in the context of interconnection networks?
Bandwidth is the number of bits that can be transmitted per second.
What is the significance of diameter in evaluating a network?
Diameter is the maximum shortest path between any two processors, and it gives the worst-case latency.
What is the primary characteristic of a Uniform Memory Access (UMA) platform?
The time taken by a processor to access any memory in the network, global or local, is uniform.
What is the main difference between Uniform Memory Access (UMA) and Non-uniform Memory Access (NUMA)?
The number of memory controllers used.
Which type of memory access is slower, Uniform Memory Access (UMA) or Non-uniform Memory Access (NUMA)?
Uniform Memory Access (UMA)
What is the main advantage of Non-uniform Memory Access (NUMA) over Uniform Memory Access (UMA)?
It has more bandwidth.
What type of applications is Uniform Memory Access (UMA) suitable for?
General purpose and time-sharing applications
What is a characteristic of non-uniform Memory Access?
Memory access time is not equal.
What is the key characteristic of memory access time in a Uniform Memory Access (UMA) system?
It is balanced or equal.
What is a characteristic of Message-Passing Platforms?
They comprise of a set of processors and their own (exclusive) memory.
What is the main limitation of Uniform Memory Access (UMA)?
It has limited bandwidth.
What is required for message passing?
Little hardware support, other than a network.
What is the time taken to prepare a message for transmission and handle it at the sending and receiving nodes called?
Startup time (ts).
What type of applications is Non-uniform Memory Access (NUMA) suitable for?
Real-time applications
What is the time taken by the header of a message to travel between two directly connected nodes in the network called?
Per-hop time (th).
What is the time taken by each word to traverse the link called?
Per-word transfer time (tw).
What is the total time taken for a message of size m to traverse l links?
th + twm.
How does the bandwidth of a channel affect the per-word transfer time?
The per-word transfer time is equal to 1/r, where r is the channel bandwidth in words per second.
What is the primary characteristic of a Parallel Random Access Machine (PRAM)?
All processors operate synchronously under a single clock and can randomly access a large shared memory.
What is the primary function of interconnection networks in parallel computers?
To carry data between processors and to memory.
What is the difference between static and dynamic interconnection networks?
Static networks consist of point-to-point communication links among processing nodes, while dynamic networks are built using switches and communication links.
What determines the cost of a switch in an interconnection network?
The cost of a switch grows as the square of the degree of the switch (𝑁 2).
How do processors communicate with the network?
Via a network interface.
What impacts the performance of the network according to the text?
The relative speeds of the I/O and memory buses.
What is the significance of the per-hop time th in parallel computers?
It is quite small and can be ignored for most parallel algorithms.
What is the composition of a PRAM?
p processors and a global memory of unbounded size that is uniformly accessible to all processors.
This quiz covers message passing architectures, including static and dynamic interconnection networks, their characteristics, and network criteria such as latency.
Make Your Own Quizzes and Flashcards
Convert your notes into interactive study material.
Get started for free
