Workload-Conditioned Health Residual Graphs for GPU Fleets

RN-001 · Jack Shaquer · July 2026 · Draft v0.1 · open research note · not peer reviewed

Read the full paper [PDF] →

Abstract. Modern GPU fleets produce rich telemetry but still make it difficult to distinguish physical device deterioration from changes in workload or failures in the monitoring path. RN-001 proposes a typed, time-varying health residual graph whose subsystem residuals are workload-conditioned, compared across multiple reference bases, normalized by physical exposure, and coupled to a closed-loop diagnostic selector.

1. Research premise

Static thresholds and workload-blind anomaly detectors can confuse normal phase changes with hardware degradation. The note therefore treats workload confounding, observability confounding, and the gap between passive evidence and active diagnosis as separate problems that must be represented explicitly.

2. Proposed representation

Each accelerator is represented through a typed health residual graph containing accelerators, HBM stacks, interconnects, hosts, cooling domains, workload phases, telemetry sources, RAS events, and diagnostic procedures. Nodes and edges retain residual magnitude, confidence, persistence, exposure denominator, timestamp, model version, and provenance rather than collapsing evidence into one anomaly score.

3. Residual construction

Observed behavior is compared against four distinct reference bases: workload-conditioned expectation, the device's own history, an equivalently configured peer cohort, and optional physics-constrained expectations. The multi-reference structure is intended to help distinguish fleet-wide or environmental effects from device-specific deterioration.

4. Exposure-normalized health

Reliability evidence is normalized by physical exposure rather than wall-clock time. The note sketches subsystem embodiments for thermal behavior, power-performance efficiency, HBM ECC and row-remap evidence, PCIe/NVLink health, control-plane state, and observability continuity.

5. Diagnostic loop

The graph state generates candidate actions such as increasing telemetry frequency, requesting additional fields, running DCGM or vendor diagnostics, replaying a controlled workload, comparing a peer, checking cooling or power sensors, collecting logs, or draining a device. Actions are ranked by expected information gain, disruption, duration, safety, operational constraints, and cost of delay.

6. Status

This is a conceptual research note shared for open discussion and collaboration. It does not claim completed empirical validation. The proposed next step is an offline DCGM replay prototype evaluated against historical fleet telemetry, incident tickets, and RMA outcomes.

Download RN-001 [PDF]