neuron-tai/.scratch/distributed-gguf-runtime/issues/057-benchmark-batching-saturation-and-fairness.md

DGR-057: Benchmark batching saturation and fairness

• Status / triage: specification only; ready-for-agent; passes: false
• Execution mode: AFK
• Milestone: M4
• Dependencies: DGR-055, DGR-056
• Blocks (derived): DGR-061, DGR-063, DGR-070
• Labels: area:performance, area:batching, type:benchmark, priority:p1, ready-for-agent
• Evidence class: real-hardware
• Hardware: required
• Model: dense-llama+deepseek-v4
• Upstream: no

Objective / description

Fresh Ralph session: read .scratch/distributed-gguf-runtime/RALPH-CONTEXT.md, source issue .scratch/distributed-gguf-runtime/issues/057-benchmark-batching-saturation-and-fairness.md, and evidence READMEs for dependencies (DGR-055, DGR-056) before changing code. Inspect live source/tests rather than trusting legacy pass states. Objective: Identify useful concurrency and saturation points without assuming one global batch size.

Acceptance criteria

• Run controlled concurrency 1/2/4/8 and record aggregate throughput, per-request p50/p95, TTFT, queue time, occupancy, and memory.
• Measure dense and V4 lanes separately where hardware permits.
• Verify bounded fairness and no prefill/decode starvation.
• Produce machine-readable saturation recommendations consumed by capability reporting, not hardcoded constants.
• Applicable shared quality gates in prd.json pass, and the evidence handoff records exact commands/results, changed files, limitations, and dependency handoff.

Shared quality gates

• Targeted deterministic tests pass; Python changes also pass python -m compileall packages tests.
• git diff --check passes.
• Default tests are model-download-free, API-credit-free, and GPU-free.
• Evidence README records exact changed files, commands/results, limitations, and dependency handoff; no fabricated evidence or inherited completion credit.
• Native changes pass focused out-of-tree CMake build and CTest; patch changes verify clean apply/check/reverse against the exact llama.cpp pin.
• Runs are opt-in and record exact artifact/split hashes, runtime/upstream pin, backend/driver, hardware, network, commands, and raw metrics. Model artifacts use configured mounted-drive storage and never /home.
• Preserve existing Transformers behavior and backend-agnostic Tracker routing/load balancing/billing/relay semantics unless an explicit versioned contract says otherwise. One scoped story commit is expected during execution, but this specification-materialization change is not committed.

Evidence handoff

Write and verify .scratch/distributed-gguf-runtime/evidence/DGR-057/README.md. Until every criterion and applicable gate has real evidence, this story remains passes: false. Legacy evidence is provenance only, not completion credit.