Inference & Serving · core
Inference Cost & Latency Optimization
Time to first token, inter-token latency, throughput, tail latency, utilization, quality, and cost per request.
Mental model
Serving optimization is a constrained frontier: quality, TTFT, token latency, throughput, availability, and cost must be measured on the same representative workload.
How to study Inference Cost & Latency Optimization
Begin by restating the mental model in your own words, then connect it to a concrete system you have built or operated. Name the mechanism, the constraint it addresses, and the trade-off it introduces. Use Splitwise: Efficient Generative LLM Inference Using Phase Splitting, Fast Inference from Transformers via Speculative Decoding, MLPerf Inference Benchmark to check details, but close the source before writing your explanation. Retrieval is the learning step; rereading is only preparation.
Next, compare Inference Cost & Latency Optimization with Inference Hardware, Local & On-device Inference. Ask what changes in correctness, latency, resource use, operability, and failure recovery. Complete Design exercise: Inference Cost & Latency Optimization and preserve the command, input, output, and one failed attempt as evidence. Finish by explaining the idea without jargon to someone who has not studied the track.
Proof of understanding
- Explain the mechanism from first principles and identify the state it reads or changes.
- Give one situation where the concept is the right choice and one where it is not.
- Predict a realistic failure mode before running the drill, then compare the prediction with evidence.
- Connect the result to a roadmap or build artifact instead of treating the concept as isolated trivia.
Learn from primary sources
- Splitwise: Efficient Generative LLM Inference Using Phase Splitting (doc)
- Fast Inference from Transformers via Speculative Decoding (doc)
- MLPerf Inference Benchmark (doc)
- DistServe: Disaggregating Prefill and Decoding for Goodput-optimized LLM Serving (OSDI '24) (paper)
- Hugging Face — LLM Inference Optimization (doc)
Practice and explain it back
Design exercise: Inference Cost & Latency Optimization
Time to first token, inter-token latency, throughput, tail latency, utilization, quality, and cost per request. Implement designOutline() returning non-empty values for: workload, latencyBreakdown, costModel. Each value must name a concrete mechanism or decision.
Expected evidence: A design outline with workload, latencyBreakdown, costModel plus an explicit failure mode or trade-off.
Open the interactive drill →Review prompts
- Why do TTFT and inter-token latency pull the scheduler in opposite directions?
Build evidence
Synthesize: Inference & Serving
Build a production mental model for inference engines, memory, kernels, routing, hardware, and serving economics. Produce one working system, benchmark, or evidence-backed design that integrates the path.
- Implements or precisely models the core mechanisms from all three milestones
- Includes at least one injected failure or adversarial case and demonstrates recovery
- Reports quality, latency, resource, reliability, or usability measurements relevant to the domain
- Ships a concise architecture note explaining decisions, trade-offs, and remaining risks