A "roofline analysis" reveals that LLM performance is limited by the slower of two factors: the time it takes to fetch model parameters from memory (memory-bound) or the time it takes to perform matrix multiplications (compute-bound). Optimizing performance requires identifying and addressing the correct bottleneck.

AI workloads are limited by memory bandwidth, not capacity. While commodity DRAM offers more bits per wafer, its bandwidth is over an order of magnitude lower than specialized HBM. This speed difference would starve the GPU's compute cores, making the extra capacity useless and creating a massive performance bottleneck.

Unlike simple classification (one pass), generative AI performs recursive inference. Each new token (word, pixel) requires a full pass through the model, turning a single prompt into a series of demanding computations. This makes inference a major, ongoing driver of GPU demand, rivaling training.

The necessity of batching stems from a fundamental hardware reality: moving data is far more energy-intensive than computing with it. A single parameter's journey from on-chip SRAM to the multiplier can cost 1000x more energy than the multiplication itself. Batching amortizes this high data movement cost over many computations.

Top inference frameworks separate the prefill stage (ingesting the prompt, often compute-bound) from the decode stage (generating tokens, often memory-bound). This disaggregation allows for specialized hardware pools and scheduling for each phase, boosting overall efficiency and throughput.

For any given hardware, there is a fundamental lower bound on inference latency. This "latency floor" is the time required to load the model's total parameters from memory (e.g., HBM) onto the chip. This process cannot be sped up by reducing batch size or other software tricks.

Traditional ML used "micro-batching" by normalizing inputs to the same size. LLMs break this model due to variable input/output lengths. The core innovation is continuous processing, handling one token at a time across all active requests, which creates complex scheduling and memory challenges solved by techniques like PagedAttention.

When splitting jobs across thousands of GPUs, inconsistent communication times (jitter) create bottlenecks, forcing the use of fewer GPUs. A network with predictable, uniform latency enables far greater parallelization and overall cluster efficiency, making it more important than raw 'hero number' bandwidth.

API providers offer faster inference at a premium by reducing the number of users processed simultaneously (batch size). This lowers latency but makes each token more expensive because the fixed cost of loading model weights is spread across fewer requests, reducing amortization.