The Workload Class Problem: Coherence-SLA Classification as a First-Class Scheduling Primitive at the Tactical Edge

The preceding four Wave 1 papers have specified the tactical substrate's mechanical components: DTN custody transfer (P3), WAL + CRDT state coherence (P4), the AI Supervisor out-of-path control plane (P5), and governed dynamic protocol selection (P6). Each of these components requires a key input that none of them generates: the workload class of the data being processed. P3's custody scheduler prioritizes bundles by workload class. P4's CRDT merge engine escalates or resolves conflicts based on workload class. P5's Supervisor applies autonomy envelopes that are workload-class-dependent. P6's protocol selection decision matrix uses workload class requirements as its primary constraint input. The workload class is the common scheduling primitive that makes the four substrate components governable rather than technically capable without direction. This paper argues that the workload class problem — how to define, assign, enforce, and evolve a taxonomy of operational data types that the substrate can use as a scheduling primitive — is not a data modeling problem but a governance architecture problem. The taxonomy is not derivable from the technical properties of the data; it is an encoding of operational doctrine and coherence requirements that only the H half can supply. A workload class taxonomy authored by an engineering team reflects engineering priorities; a workload class taxonomy authored by the operational community reflects operational doctrine. The substrate enforces whichever taxonomy it receives; its enforcement does not validate the taxonomy's operational correctness — only governance review can do that. The paper develops a four-dimension workload class framework: coherence class (what consistency guarantee does this data type require across disconnected nodes?), SLA class (what latency bound and delivery confirmation does this data type require?), security class (what authentication and encryption profile does this data type require?), and authority class (what level of governance oversight does the processing of this data type require?). Each dimension is independently assigned per data type; the combination of the four dimensions produces the complete workload class specification that the substrate uses as its scheduling primitive. The §6 Governor Application specifies how the HGC³AE² framework governs the taxonomy's definition, maintenance, and enforcement. This is Paper 7 of The Implications of Edge Degraded Ops — an 11-paper undecalogy on distributed state at the C5ISR edge under DDIL conditions. The frame paper is The Tactical Substrate; the load-bearing governance framework is HGC³AE² at the Degraded Edge. Rights envelope: Citation permitted with full attribution. No reproduction, redistribution, or derivative works without written permission. AI/ML training use disallowed. See the citation policy at https://nonsequitur.tech/pubs/citation-policy/ for the full rights envelope. Canonical site URL: https://nonsequitur.tech/white-papers/workload-class-problem/