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": 0.75,
|
|
"cell": "tp2_mns64",
|
|
"early_stop_reason": "slo_pass_rate_unrecoverable",
|
|
"early_stopped": true,
|
|
"exact_output_count": 129,
|
|
"feasible": false,
|
|
"interval": {
|
|
"elapsed_s": 26.447176041,
|
|
"end_mono_ns": 201148848516622,
|
|
"end_wall_ns": 1783869575974943780,
|
|
"start_mono_ns": 201122401340581,
|
|
"start_wall_ns": 1783869549527767999
|
|
},
|
|
"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": 64,
|
|
"observed_count": 391,
|
|
"pass_rate": 0.14066496163682865,
|
|
"schema": 1,
|
|
"selection": {
|
|
"arrival_order_sha256": "d3ed714128e985d0e0a0ca6b3767210c181114ab6694fb066a60aecf5afa9f9a",
|
|
"arrival_s": {
|
|
"distinct_n": 391,
|
|
"finite_n": 391,
|
|
"max": 59.91880000000001,
|
|
"min": 0.008300000000008368,
|
|
"missing_n": 0,
|
|
"n": 391
|
|
},
|
|
"count": 391,
|
|
"long_gt4096": 140,
|
|
"offered_req_s": 6.516666666666667,
|
|
"offered_req_s_per_gpu": 3.2583333333333333,
|
|
"raw_input_tokens": {
|
|
"distinct_n": 361,
|
|
"finite_n": 391,
|
|
"max": 8149.0,
|
|
"min": 71.0,
|
|
"missing_n": 0,
|
|
"n": 391
|
|
},
|
|
"raw_length_order_sha256": "8152bbaf9f27e84e6fbd6359174d17de4b9c96ea8b8c58e3760290be3469b2a9",
|
|
"request_id_order_sha256": "4f4c792e90f5c1bf33978fea26debcadfce5f8b872a31be86a88a1bf6b09a21c"
|
|
},
|
|
"slo_pass_count": 55,
|
|
"study_sha256": "9474f0d0b53579f1db852ca68abfb0b96ba43ae4e17738118bf8e3209eb09ece",
|
|
"tp": 2,
|
|
"tpot_ms": {
|
|
"distinct_n": 129,
|
|
"finite_n": 129,
|
|
"max": 109.8500202991083,
|
|
"min": 5.983953999894429,
|
|
"missing_n": 262,
|
|
"n": 391
|
|
},
|
|
"ttft_ms": {
|
|
"distinct_n": 129,
|
|
"finite_n": 129,
|
|
"max": 2141.4210819930304,
|
|
"min": 43.8292009930592,
|
|
"missing_n": 262,
|
|
"n": 391
|
|
}
|
|
}
|