Quick Revision

GK One-Line Question & Answer

15541+ short questions with short answers, covering every category and sub-category on the site — no long articles to scroll through. Good for a fast recap before an exam, or a few minutes of daily practice.

Looking for exam process/strategy questions instead? Try the Exam Preparation FAQ →

Computer Fundamentals → Introduction to Computer 35

Flynn's taxonomy classifies computer architectures based on?
Instruction and data stream multiplicity (SISD, SIMD, MISD, MIMD)
click to copy
GPU computing is classified as?
SIMD (Single Instruction Multiple Data)
click to copy
SIMD instructions like AVX allow CPU to?
Process multiple data elements with one instruction
click to copy
Amdahl's Law states that speedup is limited by?
The sequential (non-parallelizable) portion of the program
click to copy
Gustafson's Law (scaled speedup) argues that?
Larger problems can be solved in same time with more cores — overcomes Amdahl's limit
click to copy
Interconnect topology in supercomputers affects?
Communication latency and bandwidth between nodes
click to copy
Which interconnect technology is used in HPC clusters?
InfiniBand (low latency, high bandwidth)
click to copy
GPGPUs (General Purpose GPU) are used for?
Scientific computing, AI, video processing using massively parallel architecture
click to copy
CUDA is NVIDIA's platform for?
GPU parallel computing using C/C++ extensions
click to copy
OpenCL is?
Cross-platform parallel computing framework for GPU/CPU/FPGA
click to copy
TPU (Tensor Processing Unit) developed by Google is optimized for?
Machine learning tensor operations (matrix multiplications)
click to copy
Neuromorphic computing mimics?
Biological neural networks for energy-efficient AI
click to copy
Memory-centric computing (Processing-in-Memory) places computation?
Near or within the memory chip to reduce data movement
click to copy
Non-von Neumann architectures include?
Dataflow and neuromorphic architectures
click to copy
Dataflow architecture executes instructions when?
All required input data is available — data-driven execution
click to copy
Optical computing uses?
Photons (light) for computation — speed of light, no electrical resistance
click to copy
DNA computing uses?
DNA molecules to perform parallel computation
click to copy
Cache replacement policy LFU (Least Frequently Used) evicts?
Item accessed least number of times
click to copy
Cache thrashing (high miss rate) is caused by?
Working set larger than available cache causing frequent evictions
click to copy
Non-temporal store instruction bypasses?
Cache — writes directly to RAM for large streaming writes
click to copy
Prefetching in CPUs attempts to?
Load data into cache before it's needed based on access patterns
click to copy
Memory bandwidth vs latency: bandwidth measures?
Total data transfer rate (GB/s)
click to copy
Which memory technology uses phase change (crystalline vs amorphous)?
Phase Change Memory (PCM/3D XPoint)
click to copy
Compute Express Link (CXL) is?
High-speed interconnect enabling cache-coherent CPU-accelerator-memory communication
click to copy
Die stacking (3D IC) in chip design?
Stacks multiple chips vertically with through-silicon vias (TSV) for density
click to copy
Chiplet design disaggregates?
Monolithic chip into smaller, specialized dies connected via interconnect
click to copy
AMD EPYC processors use chiplet design with?
Multiple CPU chiplets + I/O die connected via Infinity Fabric
click to copy
EUV (Extreme Ultraviolet) lithography uses light at wavelength of?
13.5 nm
click to copy
FinFET transistor structure has?
Vertical fin-shaped channel for better gate control
click to copy
GAA (Gate All Around) FET improves on FinFET by?
Surrounding the channel on all four sides for better control at smaller nodes
click to copy
Process node (e.g., 7nm) in chip manufacturing refers to?
Marketing term roughly corresponding to transistor density
click to copy
Photolithography in chip manufacturing?
Uses light to transfer circuit patterns from mask to wafer
click to copy
Which company manufactures chips for Apple Silicon?
TSMC (Taiwan Semiconductor Manufacturing Company)
click to copy
Moore's Law is expected to slow down because?
Physical limits of silicon at atomic scale
click to copy
Beyond Moore's Law innovations include?
Chiplets, 3D stacking, new materials (GaN, SiC, 2D materials), neuromorphic
click to copy

Computer Fundamentals → Memory Units 4

FPGA (Field Programmable Gate Array) differs from CPU/GPU in?
Its digital logic can be reconfigured after manufacturing
click to copy
Resistive RAM (ReRAM/RRAM) stores data using?
Variable resistance of a dielectric material
click to copy
Magnetoresistive RAM (MRAM) uses?
Magnetic tunnel junction resistance for non-volatile storage
click to copy
Ferroelectric RAM (FeRAM) stores data using?
Polarization of ferroelectric material (non-volatile, fast, low power)
click to copy

Computer Fundamentals → Generations of Computer 1

ASIC (Application Specific Integrated Circuit) vs FPGA: ASIC is?
Fixed-function chip designed for one specific application — faster and more efficient
click to copy