Elastic Gang: Per-Token Membership Change for a Hard-Barriered LLM Inference Gang Co-Scheduled with OS Processes

https://arxiv.org/abs/2607.04668v1

Core Idea

The problem is that on-device LLM decoding is a hard-barriered CPU-SIMD computation requiring all cores per token, but preemptive scheduling causes deadlock or silent logit corruption.

For this daily profile, it is worth opening because it links Inference, LLM, and Scheduling to a concrete method, not just a broad trend.

What Is New

The novelty signal is concentrated around Inference, LLM, and Scheduling. For this profile, the important question is whether the paper changes how architecture ideas are generated, evaluated, or connected to software and hardware constraints.

Methodology

Read this as a loop: define the target system, apply the proposed mechanism, measure against a baseline, then use the measured signal to justify the next design choice. Mechanism: On-device LLM decoding is a hard-barriered CPU-SIMD computation that wants every core for milliseconds per token, while the rest of the OS wants those same cores continuously. Evidence: A barriered gang cannot simply be dropped into a preemptive scheduler: an unannounced departure deadlocks a barrier, and an unannounced arrival silently corrupts logits.

score(design) = quality_metric(design) - cost_to_evaluate(design) + feedback_gain(design)

Figure To Read First

Read this visual first: focus on the first architecture, workflow, or pipeline figure before the experiments. It should show what is optimized, what feedback signal is used, and where the system boundary sits.

Minimal Mental Model

research artifact
  question      -> what design, runtime, or system boundary changes?
  mechanism     -> model, agent, compiler, simulator, or hardware feedback
  evaluation    -> baseline comparison plus cost / latency / accuracy signal
  reusable idea -> what should carry into the next architecture experiment?

Why It Matters

Paper recommendations matter when they sharpen the research map: what problem is now easier to study, what methodology becomes reusable, and which architecture assumptions should be questioned next.