Skip to main content
← Lab

stragglerlat

confirmedPython · llama.cppInference & servingJuly 7, 2026

the question

Do long requests punish short ones under continuous batching?

what came out

A straggler taxes short-request latency (1.5x) via slot contention, not head-of-line blocking (shorts still finish first) -- but NOT more than an equal count of short requests: the matched control refutes the intuition. Real llama.cpp, pre-registered, code-reviewed.

method & receipts

  • Result: confirmed
  • Pre-registered the prediction was written down and committed before the run.
  • Reproducible one script re-runs the whole thing from scratch.
  • Independently reviewed an adversarial review pass, kept in the repo.
  • Machine-readable claims every reported number, in a checkable file.
  • Tested — a correctness/benchmark suite ships alongside the code.

→ read the code and re-run it

github.com/v-code01/stragglerlat