Operational scaling characteristics

Determine how the Brain API control plane behaves under increasing intent throughput, capability-registry scale, and active-policy-set scale, including the point at which decision latency exceeds the target operational threshold.

Background

The paper's experiments use small corpora and explicitly acknowledge that decision latency under load is unquantified. It therefore leaves the operational envelope of a Brain API deployment unresolved.

The proposed study would measure scaling independently and jointly across request throughput, registry size, and policy-set size, identifying when horizontal scaling, caching, or architectural changes are required.

References

EQ7 --- Scaling characteristics. How does the control plane behave under increasing load across three dimensions: (a) intent throughput---requests per second at which decision latency degrades beyond the EQ1 threshold; (b) registry scale---number of registered capabilities at which candidate generation becomes a bottleneck; and (c) policy set scale---number of active policies at which evaluation cost dominates the decision cycle. Scaling curves along each dimension, measured independently and jointly, characterize the operational envelope of a Brain API deployment and identify where horizontal scaling, caching, or architectural changes are required.

— Brain API: An Intent-Aware Control Plane for Policy-Governed Agentic Systems  (2609.21299 - Chernov, 18 Sep 2026) in Section 12.7, Evaluation Criteria (EQ7)