gahow/aituner
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add OpProf campaign: protocols, results, patches, run evidence (P0-P6)

Browse Source 
Workload-conditioned operator profiling on patched vLLM 0.24.0 +
Qwen3-30B-A3B/H20. H1b PASS (irregular patterns carry +23-45pp R64
raggedness, 8-45% token-efficiency loss vs rectangular controls);
mechanism decomposition kills the padding narrative and finds the
arrival-uniformization artifact (-12.9%); cross-version churn surface
shows TP2/MNS64 -29.4% across vLLM 0.20->0.24 while the argmax held.
Raw Layer-1 JSONL streams (507 MB) stay on disk, git-ignored; footer
sidecars and metrics are tracked.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Gahow Wang
2026-07-13 11:06:10 +08:00
parent 607e88da3c
commit d5b276180d
412 changed files with 125056 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

214
runs/opprof-phase6/phase6/controller-state.json Normal file
View File

@@ -0,0 +1,214 @@
{
"completed_confirmations": 3,
"completed_primary_anchors": 21,
"failures": [
{
"failure": "RuntimeError('outside GPU processes detected: [2635013, 2635031, 2635032, 2635034]')",
"wave": "W1-tp1"
}
],
"gpu_hours_total": 2.2911728494699863,
"prior_budget_stop": {
"reason": "superseded by A-P6-1 W1 acceptance",
"spent_h20_hours": 0.4323110630779864
},
"repairs": [
{
"at": 1783867821,
"id": "phase6-owned-pgid-monitor",
"prior_gpu_hours": 0.13291816790898642,
"reason": "EngineCore marker unavailable via proc environ; use frozen Phase-3 process-group ownership"
},
{
"at": 1783869388,
"fallback_cap_activated": false,
"id": "A-P6-1",
"reason": "W1 checkpoint sidecars balance; accept eight anchors without rerun"
}
],
"schema": 1,
"started_at": 1783867644.5549395,
"status": "budget_projection_stop_before_W4",
"unconfirmed_triggers": [
{
"anchor": 0.033182214016,
"cell": "tp4_mns16"
}
],
"waves": {
"W1-tp1": {
"accounting_mode": "A-P6-1-checkpoint-sidecar",
"estimate_h20_hours": 0.35,
"gpu_hours": 0.299392895169,
"primary_anchors": 8,
"readjudication_sha256": "4ce0072dc947e492f2c7eb96c9db57c68354c12e96aa4ecb53358b836e71ceb5",
"status": "complete"
},
"W2-tp2": {
"completed_at": 1783869815.3725514,
"estimate_h20_hours": 0.65,
"gpu_hours": 0.9203650120894115,
"primary_anchors": 8,
"started_at": 1783869389.3392105,
"status": "complete",
"validations": [
{
"accounting_mode": "graceful-footer",
"cell": "tp2_mns8",
"invariants": {
"anchor_intervals_present": true,
"compile_capture_pre_ready": true,
"encoded_balanced": true,
"footer_sidecar_agrees": true,
"last_step_matches": true,
"layer1_contiguous": true,
"layer1_zero_drops": true,
"one_footer_last": true,
"one_ready_marker": true,
"sidecar_final": true,
"warmup_exact_16": true,
"warmup_long": true,
"written_matches_records": true
},
"layer1_records": 6687,
"post_ready_capture_events": [],
"stream": "/home/admin/cpfs/wjh/opprof-phase6-dash0-20260712/runs/phase6/cells/tp2_mns8/opprof/opprof-v1-dp0-pid2661256-1783869521184806299.jsonl"
},
{
"accounting_mode": "graceful-footer",
"cell": "tp2_mns16",
"invariants": {
"anchor_intervals_present": true,
"compile_capture_pre_ready": true,
"encoded_balanced": true,
"footer_sidecar_agrees": true,
"last_step_matches": true,
"layer1_contiguous": true,
"layer1_zero_drops": true,
"one_footer_last": true,
"one_ready_marker": true,
"sidecar_final": true,
"warmup_exact_16": true,
"warmup_long": true,
"written_matches_records": true
},
"layer1_records": 6313,
"post_ready_capture_events": [],
"stream": "/home/admin/cpfs/wjh/opprof-phase6-dash0-20260712/runs/phase6/cells/tp2_mns16/opprof/opprof-v1-dp0-pid2661159-1783869522696890561.jsonl"
},
{
"accounting_mode": "graceful-footer",
"cell": "tp2_mns32",
"invariants": {
"anchor_intervals_present": true,
"compile_capture_pre_ready": true,
"encoded_balanced": true,
"footer_sidecar_agrees": true,
"last_step_matches": true,
"layer1_contiguous": true,
"layer1_zero_drops": true,
"one_footer_last": true,
"one_ready_marker": true,
"sidecar_final": true,
"warmup_exact_16": true,
"warmup_long": true,
"written_matches_records": true
},
"layer1_records": 17712,
"post_ready_capture_events": [],
"stream": "/home/admin/cpfs/wjh/opprof-phase6-dash0-20260712/runs/phase6/cells/tp2_mns32/opprof/opprof-v1-dp0-pid2661056-1783869524249984543.jsonl"
},
{
"accounting_mode": "graceful-footer",
"cell": "tp2_mns64",
"invariants": {
"anchor_intervals_present": true,
"compile_capture_pre_ready": true,
"encoded_balanced": true,
"footer_sidecar_agrees": true,
"last_step_matches": true,
"layer1_contiguous": true,
"layer1_zero_drops": true,
"one_footer_last": true,
"one_ready_marker": true,
"sidecar_final": true,
"warmup_exact_16": true,
"warmup_long": true,
"written_matches_records": true
},
"layer1_records": 4537,
"post_ready_capture_events": [],
"stream": "/home/admin/cpfs/wjh/opprof-phase6-dash0-20260712/runs/phase6/cells/tp2_mns64/opprof/opprof-v1-dp0-pid2661205-1783869525995600880.jsonl"
}
]
},
"W3-tp4-trap": {
"completed_at": 1783870244.6658168,
"estimate_h20_hours": 0.85,
"gpu_hours": 0.9384967743025886,
"primary_anchors": 5,
"started_at": 1783869815.6737785,
"status": "complete",
"validations": [
{
"accounting_mode": "graceful-footer",
"cell": "tp4_mns8",
"invariants": {
"anchor_intervals_present": true,
"compile_capture_pre_ready": true,
"encoded_balanced": true,
"footer_sidecar_agrees": true,
"last_step_matches": true,
"layer1_contiguous": true,
"layer1_zero_drops": true,
"one_footer_last": true,
"one_ready_marker": true,
"sidecar_final": true,
"warmup_exact_16": true,
"warmup_long": true,
"written_matches_records": true
},
"layer1_records": 10954,
"post_ready_capture_events": [],
"stream": "/home/admin/cpfs/wjh/opprof-phase6-dash0-20260712/runs/phase6/cells/tp4_mns8/opprof/opprof-v1-dp0-pid2670761-1783869954643149539.jsonl"
},
{
"accounting_mode": "graceful-footer",
"cell": "tp4_mns16",
"invariants": {
"anchor_intervals_present": true,
"compile_capture_pre_ready": true,
"encoded_balanced": true,
"footer_sidecar_agrees": true,
"last_step_matches": true,
"layer1_contiguous": true,
"layer1_zero_drops": true,
"one_footer_last": true,
"one_ready_marker": true,
"sidecar_final": true,
"warmup_exact_16": true,
"warmup_long": true,
"written_matches_records": true
},
"layer1_records": 26137,
"post_ready_capture_events": [],
"stream": "/home/admin/cpfs/wjh/opprof-phase6-dash0-20260712/runs/phase6/cells/tp4_mns16/opprof/opprof-v1-dp0-pid2670831-1783869959685052499.jsonl"
}
]
}
},
"budget_stop": {
"before_wave": "W4-tp4",
"spent_h20_hours": 2.2911728494699863,
"wave_estimate_h20_hours": 0.65,
"safety_buffer_h20_hours": 0.10,
"projected_total_h20_hours": 3.0411728494699863,
"hard_cap_h20_hours": 3.0,
"excess_h20_hours": 0.04117284946998634,
"remaining_cells": [
"tp4_mns32",
"tp4_mns64"
],
"remaining_primary_anchors": 4
}
}