Hybrid Dataflow Selection: Debug Playbook

Debug playbook

Debug Playbook for Hybrid Dataflow Selection is anchored on End-to-end joules per inference and latency improvement from per-layer dataflow switching versus single-style baseline.. Convert measurements into mechanism-backed decisions with clear owner accountability.

1. Freeze workload seed, model revision, and execution environment.

2. Locate first persistent stage where metrics diverge.

3. Classify dominant mechanism: compute, memory, scheduling, precision, or thermal.

4. Build one focused reproducer and apply one bounded fix.

5. Re-run full correctness, quality, and performance matrix.

Review memo template

ACCELERATOR REVIEW MEMO - Dataflow Architectures / Hybrid Dataflow Selection

1. Symptom
   - Failing metric: End-to-end joules per inference and latency improvement from per-layer dataflow switching versus single-style baseline.
   - Workload or traffic slice: <name>
   - First failing layer or stage: <operator, schedule, memory, runtime>
   - Build and runtime tags: <compiler/firmware/runtime/hardware>

2. Mechanism hypothesis
   - Primary mechanism: Hybrid policies choose different dataflows by operator class or shape regime instead of enforcing one stationary style globally. For example, output-stationary may win on deep reductions while weight-stationary performs better on high-filter-reuse layers, and input-stationary can help bandwidth-bound feature maps. The key is selecting switch points that account for retile overhead, control complexity, and compiler/runtime transition cost. Teams usually rely on profiling-guided heuristics or cost models that include both kernel efficiency and orchestration penalty.
   - Competing hypotheses: <dataflow mismatch, memory stalls, precision drift, thermal limits>
   - Missing evidence: <counter packet, trace, replay, signoff data>

3. Proposed action
   - Smallest reversible change: <mapping/runtime/policy/config>
   - Expected movement: <throughput, p99 latency, perf-per-watt>
   - Regression risk: correctness, quality, thermal, software compatibility

4. Signoff
   - Required artifact: Dataflow policy playbook with layer-wise recommendations, transition rules, and expected gains.
   - Required owners: system architect, compiler/runtime owner, performance modeling owner, production inference lead
   - Final decision: ship, bounded rollout, rollback, or escalate

AI accelerator deep dive

Dataflow choices are durable architecture decisions that shape memory and scheduling cost.

Concept diagram

DATAFLOW DECISION

input/output/weight stationary
  -> locality pattern
  -> movement cost
  -> throughput and power

Metric graph

DATAFLOW COST MIX

activation traffic   ███████
weight traffic       █████
partial-sum traffic  ██████

Metrics and artifacts to collect

• reuse factor map

• buffer pressure profile

• NoC traffic mix

• shape sensitivity analysis

Mini case study

A dataflow that won for convolution lost on attention-heavy batches due to activation movement pressure.

Debug branches

• Segment by model family

• Compare reuse vs movement

• Re-check mapping assumptions under batch variance

Senior review question

Ask: which first-principles bottleneck class explains the symptom, and what artifact proves it reproducibly?

Key takeaways

• Tie every accelerator claim to a reproducible workload slice and one primary metric trend.

• Prefer bounded fixes with clear owner and rollback boundary over broad tuning bundles.

Common pitfalls

• Optimizing synthetic kernels without production-shape validation.

• Reading average latency while ignoring p95 and p99 behavior.

• Declaring sparse or precision wins without fallback and quality evidence.

Principal accelerator review addendum

Hybrid Dataflow Selection should be framed as a full-system behavior, not an isolated kernel trick. Production outcomes are set by model shape mix, compiler choices, runtime queueing policy, memory hierarchy limits, and silicon delivery margins.

Hybrid policies choose different dataflows by operator class or shape regime instead of enforcing one stationary style globally. For example, output-stationary may win on deep reductions while weight-stationary performs better on high-filter-reuse layers, and input-stationary can help bandwidth-bound feature maps. The key is selecting switch points that account for retile overhead, control complexity, and compiler/runtime transition cost. Teams usually rely on profiling-guided heuristics or cost models that include both kernel efficiency and orchestration penalty. A useful explanation always ties observed symptom to a repeatable path where useful work was blocked, delayed, or diluted by overhead.

Use End-to-end joules per inference and latency improvement from per-layer dataflow switching versus single-style baseline. as an alarm, then anchor action using hard evidence such as Dataflow policy playbook with layer-wise recommendations, transition rules, and expected gains..

Dataflow selection governs reuse, movement cost, and predictability across model shapes. Senior reviews expect a chain of proof: workload intent -> mapping -> hardware behavior -> product impact.

Use this addendum to force explicit owner assignment, bounded fixes, and reproducible evidence before declaring closure.