Gahow Wang
d5b276180d
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>
75 lines
1.9 KiB
JSON
75 lines
1.9 KiB
JSON
{
|
|
"anchor": 0.75390625,
|
|
"cell": "tp2_mns32",
|
|
"early_stop_reason": "slo_pass_rate_unrecoverable",
|
|
"early_stopped": true,
|
|
"exact_output_count": 110,
|
|
"feasible": false,
|
|
"interval": {
|
|
"elapsed_s": 27.144585398,
|
|
"end_mono_ns": 201149546363473,
|
|
"end_wall_ns": 1783869576672790977,
|
|
"start_mono_ns": 201122401778075,
|
|
"start_wall_ns": 1783869549528204982
|
|
},
|
|
"invariants": {
|
|
"arrival_nondecreasing": true,
|
|
"exact_output_or_failed": true,
|
|
"outcomes_cover_selected": true,
|
|
"raw_lengths_present": true,
|
|
"selected_nonempty": true,
|
|
"warmup_16": true,
|
|
"warmup_exact_16": true,
|
|
"warmup_long": true
|
|
},
|
|
"kind": "anchor",
|
|
"mns": 32,
|
|
"observed_count": 394,
|
|
"pass_rate": 0.027918781725888325,
|
|
"schema": 1,
|
|
"selection": {
|
|
"arrival_order_sha256": "60fc43518e79fd46abe59afde67663a762ca17e817202fa52803040ac9056796",
|
|
"arrival_s": {
|
|
"distinct_n": 394,
|
|
"finite_n": 394,
|
|
"max": 59.91880000000001,
|
|
"min": 0.008300000000008368,
|
|
"missing_n": 0,
|
|
"n": 394
|
|
},
|
|
"count": 394,
|
|
"long_gt4096": 142,
|
|
"offered_req_s": 6.566666666666666,
|
|
"offered_req_s_per_gpu": 3.283333333333333,
|
|
"raw_input_tokens": {
|
|
"distinct_n": 364,
|
|
"finite_n": 394,
|
|
"max": 8149.0,
|
|
"min": 71.0,
|
|
"missing_n": 0,
|
|
"n": 394
|
|
},
|
|
"raw_length_order_sha256": "4a9cc03e351cce8c558b30732418c878fc3b6d24e64bef6717ce7c87c384384f",
|
|
"request_id_order_sha256": "2677a6a12ff0e104d6246dd121d0667ff4fe3d25f8d68699547906420246ebdd"
|
|
},
|
|
"slo_pass_count": 11,
|
|
"study_sha256": "9474f0d0b53579f1db852ca68abfb0b96ba43ae4e17738118bf8e3209eb09ece",
|
|
"tp": 2,
|
|
"tpot_ms": {
|
|
"distinct_n": 110,
|
|
"finite_n": 110,
|
|
"max": 85.24905312595537,
|
|
"min": 5.966831590589796,
|
|
"missing_n": 284,
|
|
"n": 394
|
|
},
|
|
"ttft_ms": {
|
|
"distinct_n": 110,
|
|
"finite_n": 110,
|
|
"max": 8747.10591501207,
|
|
"min": 43.805207998957485,
|
|
"missing_n": 284,
|
|
"n": 394
|
|
}
|
|
}
|