NCP-GENL GPU Acceleration and Optimization Practice Question
When deploying a large language model on NVIDIA H100 GPUs using TensorRT-LLM, which TWO configuration strategies are most effective for improving KV cache efficiency and memory utilization?
⚠ Common exam trap
Candidates frequently select generic training optimizations like data parallelism instead of focusing specifically on inference-centric KV cache and batching strategies.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Enable PagedAttention to reduce fragmentation.
Optimizing the KV cache is critical for scaling generative AI, as it occupies significant VRAM during inference. PagedAttention dynamically manages memory blocks to prevent fragmentation, while inflight batching ensures that continuous requests are processed without waiting for the entire batch to finish. Combining these allows for higher concurrency and reduced memory overhead, enabling more efficient deployment of models with large context windows on limited GPU hardware.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Enable PagedAttention to reduce fragmentation.
Why this is correct
PagedAttention treats the KV cache as non-contiguous memory blocks, similar to virtual memory in an OS. This eliminates internal fragmentation caused by over-allocating memory for sequence lengths that never materialize, allowing the GPU to pack more requests into the same VRAM capacity for higher throughput.
- ✗
Increase the static sequence length allocation.
Why it's wrong here
Static allocation of maximum sequence lengths leads to massive memory waste, as most requests use only a fraction of the allocated buffer. This approach reduces overall system capacity and limits the number of concurrent users, making it counterproductive to efficient KV cache utilization in production.
- ✓
Enable continuous inflight batching.
Why this is correct
Inflight batching allows the scheduler to inject new requests into the batch as soon as others finish, rather than waiting for the entire batch to conclude. This maximizes GPU utilization and keeps the KV cache active and efficient by maintaining a steady stream of tokens for processing.
- ✗
Disable multi-head attention optimizations.
Why it's wrong here
Multi-head attention optimizations are designed to improve compute efficiency and memory access patterns during inference. Disabling these features would result in redundant calculations and suboptimal usage of the tensor cores, leading to increased latency and significantly lower performance for high-concurrency generative AI workloads.
- ✗
Switch to FP64 precision for all tensors.
Why it's wrong here
FP64 precision is primarily intended for scientific simulation and requires vastly more memory bandwidth and storage than FP16 or BF16. Using FP64 for LLM inference would drastically reduce throughput and increase memory pressure without providing meaningful improvements in generative text quality or model accuracy.
About these practice questions
This NCP-GENL question is part of Courseiva's 352-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official NVIDIA exam blueprint
This NCP-GENL practice question is part of Courseiva's free NVIDIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the NCP-GENL exam.