Module executor 

Query executor for RIR nodes

The executor interprets RIR (Relational IR) nodes using the CUDA kernel provider to execute GPU-accelerated relational operations.

Structs

AdaptiveJoinObservation

Runtime join observation used by adaptive re-optimization decisions.

AdaptiveReoptimizationDecision

Typed adaptive re-optimization decision.

AdaptiveReoptimizationDiagnostic

Typed adaptive re-optimization diagnostic.

AdaptiveReoptimizationStats

Runtime adaptive re-optimization telemetry.

CommonSubexpressionStats

Runtime common subexpression elimination telemetry.

DeltaRecomputeStats

Runtime summary for a delta recomputation pass.

EpistemicGpuBatchExecutionResult

Results plus aggregate trace from a split/batch epistemic GPU execution.

EpistemicGpuBatchExecutionTrace

Batch-level trace proving split components reused the single-plan GPU path.

EpistemicGpuCandidateGenerationTrace

Trace proving candidate assumptions were generated by a GPU kernel.

EpistemicGpuCandidateValidationTrace

Trace proving staged candidate buffers were validated by a GPU kernel.

EpistemicGpuConstraintValidationTrace

Bounded validation of reduced integrity-constraint relations after GPU execution.

EpistemicGpuConstraintWorldViewValidationTrace

Trace proving epistemic integrity constraints were evaluated against world views on GPU.

EpistemicGpuExecutionResult

Output from executing the reduced production runtime plan for an epistemic program.

EpistemicGpuFinalResultMaterializationTrace

Trace proving final result flags were materialized from device-side output metadata.

EpistemicGpuFinalResultTransferTrace

Trace accounting for the bounded final-result transfer after the GPU hot path.

EpistemicGpuFinalTupleMaterializationTrace

Trace proving final query tuples were materialized into a device-resident buffer.

EpistemicGpuKernelTimingTrace

CUDA-event timing captured around one epistemic GPU kernel launch.

EpistemicGpuMaterializationTrace

Trace proving accepted-candidate materialization staging used a GPU kernel.

EpistemicGpuModelMembershipTrace

Trace proving model-membership staging was performed by a GPU kernel.

EpistemicGpuPreparedExecution

Prepared runtime state for epistemic GPU execution.

EpistemicGpuPropagationTrace

Trace proving candidate propagation staging was performed by a GPU kernel.

EpistemicGpuProviderIdentity

CUDA provider identity that produced an epistemic GPU execution result.

EpistemicGpuRuntimeCounters

Runtime counters relevant to epistemic GPU certification.

EpistemicGpuRuntimePreflight

Runtime preflight summary for an epistemic executable plan.

EpistemicGpuRuntimeTrace

Counter trace captured around a reduced production runtime dispatch.

EpistemicGpuTransferBudgetTrace

Trace proving the epistemic GPU hot path avoided tracked data-plane host transfers.

EpistemicGpuWorkspace

Device-resident buffers for epistemic Generate-Propagate-Test execution.

EpistemicGpuWorkspaceCapacities

Capacity limits for an epistemic GPU workspace allocation.

EpistemicGpuWorkspaceLayout

Concrete device-buffer layout for an epistemic GPU workspace.

EpistemicGpuWorkspaceResetTrace

Trace proving an epistemic GPU workspace was initialized on device.

EpistemicGpuWorldViewValidationTrace

Trace proving staged model memberships were validated against world views on GPU.

Executor

Query executor that interprets RIR nodes using GPU kernels

JoinIndexCacheStats

Persistent join-index manager telemetry.

RelationDelta

Incremental update for a base relation.

Enums

AdaptiveReoptimizationAction

Deterministic adaptive re-optimization decision action.

AdaptiveReoptimizationDiagnosticKind

Typed adaptive re-optimization diagnostic kind.

EpistemicGpuModelMembershipSource

Source of GPU model-membership bytes for epistemic world-view validation.

EpistemicGpuRejectionReason

Typed interpretation of nonzero GPU epistemic rejection codes.

EpistemicGpuRuntimeWcojCertification

WCOJ certification status for an epistemic runtime dispatch attempt.