teleo-codex/inbox/archive/2016-00-00-corless-aimd-dynamics-distributed-resource-allocation.md

type	title	author	url	date	domain	format	status	tags	processed_by	processed_date	claims_extracted	enrichments_applied	extraction_model	extraction_notes
source	AIMD Dynamics and Distributed Resource Allocation	Martin J. Corless, C. King, R. Shorten, F. Wirth (SIAM)	https://epubs.siam.org/doi/book/10.1137/1.9781611974225	2016-01-01	internet-finance	paper	processed	pipeline-architecture operations-research AIMD distributed-resource-allocation congestion-control fairness	rio	2026-03-11	aimd-converges-to-fair-resource-allocation-without-global-coordination-through-local-congestion-signals.md reactive-control-from-observed-state-outperforms-predictive-optimization-in-variable-load-systems.md	coordination-mechanisms.md optimal-governance-requires-mixing-mechanisms-because-different-decisions-have-different-manipulation-risk-profiles.md	anthropic/claude-sonnet-4.5	Extracted two mechanism-design claims about AIMD as a distributed coordination primitive. The source is a formal mathematical treatment (SIAM monograph) providing proven properties rather than empirical observations, justifying 'proven' confidence for the core AIMD claim. The application to pipeline architecture mentioned in the source notes is treated as an extension/implication rather than a separate claim, since it's a direct application of the proven AIMD properties. Enriched existing coordination mechanism claims with AIMD as a concrete example of domain-specific mechanism selection.

AIMD Dynamics and Distributed Resource Allocation

SIAM monograph on AIMD (Additive Increase Multiplicative Decrease) as a general-purpose distributed resource allocation mechanism. Extends the TCP congestion control principle to resource allocation in computing, energy, and other domains.

Key Content

• AIMD is the most widely used method for allocating limited resources among competing agents without centralized control
• Core algorithm: additive increase when no congestion (rate += α), multiplicative decrease when congestion detected (rate *= β, where 0 < β < 1)
• Provably fair: converges to equal sharing of available bandwidth/capacity
• Provably stable: system converges regardless of number of agents or parameter values
• Three sample applications: internet congestion control, smart grid energy allocation, distributed computing
• Key property: no global information needed — each agent only needs to observe local congestion signals

Relevance to Teleo Pipeline

AIMD provides a principled, proven scaling algorithm: when eval queue is shrinking (no congestion), increase extraction workers by 1 per cycle. When eval queue is growing (congestion), halve extraction workers. This doesn't require predicting load, modeling arrivals, or solving optimization problems — it reacts to observed system state and is mathematically guaranteed to converge. Perfect for our "expensive compute, variable load" setting.

Key Facts

• AIMD algorithm: additive increase (rate += α) when no congestion, multiplicative decrease (rate *= β, 0 < β < 1) when congestion detected
• AIMD is the most widely used distributed resource allocation method (TCP congestion control)
• AIMD provably converges to equal sharing regardless of agent count or parameter values
• AIMD requires only local congestion signals, no global information