Question 1

An attractive feature of an SMP is that the existence of multiple processors is transparent to the user.

Accepted Answer

An SMP (Symmetric multiprocessing) architecture allows multiple processors to work together as a single system, with the advantage of improved performance and reliability. The user is not aware of the number of processors involved in the system, as the operating system distributes the workload among them automatically, making it transparent to the user.

Question 2

SMPs,clusters,and NUMA systems fit into the ________ category of computer systems.

Accepted Answer

SMPs (Symmetric Multi-Processing), clusters, and NUMA (Non-Uniform Memory Access) systems all involve multiple processors or nodes working together in parallel to execute tasks, which falls under the category of MIMD (Multiple Instruction, Multiple Data) systems. SISD (Single Instruction, Single Data) systems involve a single processor executing tasks sequentially, SIMD (Single Instruction, Multiple Data) systems involve a single processor executing the same instruction on multiple data simultaneously, and MISD (Multiple Instruction, Single Data) systems involve multiple processors executing different instructions on the same data simultaneously, but these do not apply to SMPs, clusters, or NUMA systems.

Question 3

Software cache coherence schemes attempt to avoid the need for additional hardware circuitry and logic by relying on the compiler and operating system to deal with the problem.

Accepted Answer

Software cache coherence schemes indeed rely on the compiler and operating system to manage cache coherence, avoiding the need for specialized hardware mechanisms.

Question 4

Both clusters and symmetric multiprocessors provide a configuration with multiple processors to support high-demand applications.

Accepted Answer

Both clusters and symmetric multiprocessors employ multiple processors to handle high-demand applications.

Question 5

With a write-update protocol there can be multiple readers but only one writer at a time.

Accepted Answer

The answer of With a write-update protocol there can be...

Question 6

A cloud broker is useful when cloud services are too complex for a cloud consumer to easily manage.

Accepted Answer

A cloud broker acts as an intermediary between cloud providers and cloud consumers, helping the latter navigate the complexities of cloud computing environments, such as selecting the most appropriate cloud services, provisioning resources, managing costs, and ensuring compliance and security. Therefore, a cloud broker is useful when cloud services are too complex for a cloud consumer to easily manage.

Question 7

With multithreading the instruction stream is divided into several smaller streams,known as threads,such that the threads can be executed in parallel.

Accepted Answer

Multithreading allows for the simultaneous execution of multiple threads within a single process, enabling parallel processing and improved performance. Each thread is an independent sequence of instructions that can be executed concurrently with other threads, enabling more efficient use of system resources and shorter processing times.

Question 8

A taxonomy first introduced by _______ is still the most common way of categorizing systems with parallel processing capability.

Accepted Answer

The taxonomy first introduced by Michael Flynn in 1966 is still the most widely used way of categorizing systems with parallel processing capability. The taxonomy categorizes systems based on the number of instruction streams and data streams involved in parallel processing. It includes four categories: SISD (single instruction stream, single data stream), SIMD (single instruction stream, multiple data streams), MISD (multiple instruction streams, single data stream), and MIMD (multiple instruction streams, multiple data streams).

Question 9

An L1 cache that does not connect directly to the bus cannot engage in a snoopy protocol.

Accepted Answer

A snoopy protocol requires direct access to the system bus in order to monitor and handle cache coherence. A cache that is not connected directly to the bus, such as an L1 cache that is embedded in a processor core, cannot engage in a snoopy protocol. Instead, it typically uses a directory-based protocol to manage coherence with other caches in the system.

Question 10

An SMP operating system manages processor and other computer resources so that the user perceives a single operating system controlling system resources.

Accepted Answer

SMP (Symmetrical Multiprocessing) operating systems are designed to manage multiple processor cores and other resources in a way that presents a single, unified operating system to the user. This allows for efficient use of system resources and improved performance for multi-threaded applications. So the statement is true.

Question 11

Vector and array processors fall into the ________ category of computer systems.

Accepted Answer

Vector and array processors utilize Single Instruction Multiple Data (SIMD) architecture which allows the processor to perform the same operation on multiple pieces of data simultaneously. This is in contrast to SISD (Single Instruction Single Data) architecture which processes one data item at a time, MISD (Multiple Instruction Single Data) which processes one data item with multiple instructions, and MIMD (Multiple Instruction Multiple Data) architecture which processes multiple data items with multiple instructions concurrently.

Question 12

The main drawback of the bus organization is reliability.

Accepted Answer

The main drawback of the bus organization is not reliability but rather its limited bandwidth and potential for data bottlenecks, as all components share the same communication line.

Question 13

Symmetric multiprocessors (SMPs)are one of the earliest,and still the most common,example of parallel organization.

Accepted Answer

SMPs are indeed one of the earliest and most common examples of parallel organization.

Question 14

The function of switching applications and data resources over from a failed system to an alternative system in the cluster is referred to as failback.

Accepted Answer

The function of switching applications and data resources over from a failed system to an alternative system in the cluster is referred to as failover. Failback refers to the process of switching back to the original system after it has been repaired or restored.

Question 15

A _________ problem arises when multiple copies of the same data can exist in different caches simultaneously,and if processors are allowed to update their own copies freely,an inconsistent view of memory can result.

Accepted Answer

The problem described in the statement is a cache coherence problem, which arises when multiple copies of the same data can exist in different caches simultaneously, and if processors are allowed to update their own copies freely, an inconsistent view of memory can result. The other options, cluster, failover, and failback, do not describe this specific problem.

Question 16

The term SMP refers to a computer hardware architecture and also to the operating system behavior that reflects that architecture.

Accepted Answer

SMP (Symmetric Multiprocessing) refers to both the hardware architecture and the operating system behavior that reflects that architecture, which allows multiple processors to share the same memory and peripherals, and work on the same task simultaneously.

Question 17

The objective with NUMA is to maintain a transparent system wide memory while permitting multiple multiprocessor nodes, each with its own bus or other internal interconnect system.

Accepted Answer

This is a correct statement about NUMA (Non-Uniform Memory Access) architecture. Its primary goal is to allow multiple CPUs or nodes with their local memory to function as a single system with a unified view of physical memory. Each CPU node can access global memory, but local memory is faster and preferred. The goal of NUMA is to provide high scalability by reducing contention and latency caused by remote memory access.

Question 18

The most important measure of performance for a processor is the rate at which it executes instructions.

Accepted Answer

The rate at which a processor executes instructions is commonly referred to as its clock speed or frequency, and it is a fundamental metric for its performance. Other factors such as the number of cores, cache size, and instruction set also contribute to overall performance but the clock speed remains a key consideration when evaluating processor performance.

Question 19

Uniprocessors fall into the _______ category of computer systems.

Accepted Answer

Uniprocessors fall into the SISD (Single Instruction Single Data) category of computer systems, which means that they execute one instruction on one piece of data at a time. A is incorrect because MIMD (Multiple Instruction Multiple Data) systems have multiple processors that can execute instructions independently. B is incorrect because SIMD (Single Instruction Multiple Data) systems have multiple processing elements that execute the same instruction on multiple pieces of data simultaneously. D is incorrect because MISD (Multiple Instruction Single Data) systems have multiple processors that execute different instructions on the same data.

Question 20

Snoopy protocols are not suitable for a bus-based multiprocessor.

Accepted Answer

Snoopy protocols are specifically designed for bus-based multiprocessor systems to maintain cache coherence by monitoring the bus to check if any shared data is being accessed or modified.

Quiz 17: Parallel Processing