Computer Architecture¶

1. In which century did the concept of parallel processing originate?

• A. 17^th century
• B. 18^th century
• C. 19^th century
• D. 20^th century

2. True or False: The Von Neumann architecture is known for efficient I/O access to memory without any bottlenecks.

3. What is the primary drawback of the Von Neumann architecture?

• A. Limited memory capacity
• B. High power consumption
• C. Von Neumann bottleneck
• D. Lack of multi-threading support

4. Which of the following benchmarks measures floating-point performance of computers?

• A. SPEC
• B. LINPACK
• C. Geekbench
• D. Cinebench

5. True or False: Parallelization within a single processor is typically achieved through multiprocessing.

Multicore CPU¶

1. According to Flynn's taxonomy, which classification is ideal for multicore CPUs that support thread-level parallelism?

• A. SISD
• B. SIMD
• C. MISD
• D. MIMD

2. True or False: In multicore CPUs, each core generally has its own L2 cache and shares an L3 cache with other cores.

3. Which architecture under Flynn's taxonomy achieves data-level parallelism by applying the same instruction to multiple data?

• A. SISD
• B. SIMD
• C. MISD
• D. MIMD

4. What is the primary purpose of the memory hierarchy in CPUs?

• A. To increase the number of cores
• B. To decrease the number of instructions
• C. To optimize access times to frequently used data
• D. To enable multiprocessing

5. True or False: MISD (Multiple Instruction Single Data) architecture is commonly used in general-purpose computing.

Shared Memory Architectures¶

1. Which shared memory architecture has uniform memory access times across all processors?

• A. Distributed Shared Memory
• B. Non-Uniform Memory Access (NUMA)
• C. Uniform Memory Access (UMA)
• D. None of the above

2. True or False: In Non-Uniform Memory Access (NUMA), each processor has faster access to its local memory segment compared to remote segments.

3. Distributed Shared Memory combines elements of which two architectures?

• A. UMA and NUMA
• B. SIMD and MIMD
• C. SISD and MISD
• D. None of the above

4. Which of the following is also known as Symmetric Multi-Processing?

• A. NUMA
• B. UMA
• C. Distributed Memory
• D. SIMD

5. True or False: Shared-memory architectures are best suited for applications that require distributed data processing across multiple locations. Answer: False

Distributed Memory Architectures¶

1. In distributed memory architecture, how do nodes typically communicate with each other?

• A. Through shared memory space
• B. By message-passing protocols
• C. Using SIMD instructions
• D. Directly through each CPU core

2. True or False: The Fat-Tree topology is popular in distributed memory architectures because of its versatility and high bandwidth.

3. Which of the following is a widely used network topology in distributed memory supercomputers?

• A. Full Mesh
• B. Star
• C. Fat-Tree
• D. Ring

4. Which programming model is commonly used to develop applications for distributed memory systems?

• A. CUDA
• B. MPI
• C. OpenMP
• D. POSIX

5. True or False: Distributed memory architectures rely on a single global memory that all processors can access directly.