Gahow Wang
90127c3389
Sweep on dash1 GPU 0 → dash2 GPU 0 over 200 Gbps RoCE. remote_bootstrap_addr=http://172.27.123.142:8998. Same 9-size × 5-rep config as the 2026-05-27 intra-node run. Per-size pure_transfer (p50) lines up within 1–3% of the intra-node numbers across all sizes: size intra p50 inter p50 512 tok 5.3 ms 5.2 ms 2048 tok 20.6 20.0 8192 tok 83.7 80.9 32k tok 320.9 309.6 64k tok 1895 1734 (bimodal in both) 128k tok 2835 2818 (bimodal in both) => Mooncake's batch_transfer_sync_write **does not use NVLink** for intra-node peers; both paths go through the 200 Gbps RDMA NIC, with the 200 Gbps NIC (not the GPU interconnect) being the bottleneck. The ~9.7 GB/s steady-state ceiling and the 6+ GiB variance regime are identical across topologies. Operational implication for §3.2: PD-disaggregation does not get cheaper by co-locating P and D on the same node — every routed request pays the same ~10 GB/s ceiling for KV transfer, no matter where it lands. Halving the transfer cost cannot be bought back by topology. Caveat: B's receive_kv events did not log on dash2 — `MB2_LOG_DIR` env var did not propagate through vLLM's EngineCore subprocess on the consumer host (cat /proc/$ENGINE_PID/environ is empty on dash2 for that var, but the producer host on dash1 worked). For this run pure_transfer numbers are from A's send_blocks alone; full rx_total breakdown is not available, but pure_transfer is the dominant term. Adds: - analyze_mb2_send_only.py — analyzer that works from A's send_blocks alone when B's receive_kv events are absent - plot_mb2_compare.py — overlay intra vs inter on the same axes - plot_mb2.py — tolerate the `rows`-less send-only schema - figs/mb2_transfer_{time,bw}_inter.png — inter-node single-curve - figs/mb2_transfer_{time,bw}_compare.png — intra vs inter overlay - analysis/mb2/A_inter_kvboth.jsonl, inter_kvboth_client.json, inter_kvboth_breakdown.json - analysis/mb2/README.md — Summary block updated to reference both paths, dated 2026-05-27 run-log entry appended with the full table and the topology-independence framing Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
679 lines
22 KiB
JSON
679 lines
22 KiB
JSON
{
|
|
"model": "/home/admin/cpfs/wjh/models/Qwen/Qwen3-Coder-30B-A3B-Instruct",
|
|
"kv_bytes_per_token": 98304,
|
|
"src_host": "172.27.123.142",
|
|
"src_port": 8000,
|
|
"dst_host": "172.27.123.133",
|
|
"dst_port": 8001,
|
|
"config_label": "inter-kvboth",
|
|
"raw": [
|
|
{
|
|
"input_tokens": 512,
|
|
"session": "677ca02530774bae990d549d0591d336",
|
|
"t_step1_client_unix": 1779885642.2549865,
|
|
"t_step2_client_unix": 1779885642.3058398,
|
|
"t_step2_end_unix": 1779885642.3249059,
|
|
"t_prefill_s": 0.05082751903682947,
|
|
"t_transfer_s": 0.019030610972549766,
|
|
"t_followup_s": 0.01642513304250315,
|
|
"cached_followup": 496,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 512,
|
|
"session": "6758b8079675495a9a13a73db1199a71",
|
|
"t_step1_client_unix": 1779885642.3416288,
|
|
"t_step2_client_unix": 1779885642.3896646,
|
|
"t_step2_end_unix": 1779885642.40659,
|
|
"t_prefill_s": 0.04800783301470801,
|
|
"t_transfer_s": 0.016899197013117373,
|
|
"t_followup_s": 0.017115428985562176,
|
|
"cached_followup": 496,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 512,
|
|
"session": "e9da485a88cc40eb927e18c37298d815",
|
|
"t_step1_client_unix": 1779885642.4239926,
|
|
"t_step2_client_unix": 1779885642.4724562,
|
|
"t_step2_end_unix": 1779885642.492528,
|
|
"t_prefill_s": 0.04843967000488192,
|
|
"t_transfer_s": 0.02004897501319647,
|
|
"t_followup_s": 0.016289777995552868,
|
|
"cached_followup": 496,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 512,
|
|
"session": "a70db16a01b04409b04ef30e94082720",
|
|
"t_step1_client_unix": 1779885642.509085,
|
|
"t_step2_client_unix": 1779885642.5566719,
|
|
"t_step2_end_unix": 1779885642.577205,
|
|
"t_prefill_s": 0.047563806001562625,
|
|
"t_transfer_s": 0.020512140006758273,
|
|
"t_followup_s": 0.016482150997035205,
|
|
"cached_followup": 496,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 512,
|
|
"session": "d13cddd850904d4591b15c112ab32a5f",
|
|
"t_step1_client_unix": 1779885642.5939884,
|
|
"t_step2_client_unix": 1779885642.6417143,
|
|
"t_step2_end_unix": 1779885642.6617038,
|
|
"t_prefill_s": 0.04769411502638832,
|
|
"t_transfer_s": 0.019964047998655587,
|
|
"t_followup_s": 0.01518680399749428,
|
|
"cached_followup": 496,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 1024,
|
|
"session": "64d9b1335bd04698ab4f681fe4fe1e40",
|
|
"t_step1_client_unix": 1779885642.6773622,
|
|
"t_step2_client_unix": 1779885642.7512894,
|
|
"t_step2_end_unix": 1779885642.7811818,
|
|
"t_prefill_s": 0.07390057999873534,
|
|
"t_transfer_s": 0.029867444012779742,
|
|
"t_followup_s": 0.017206061049364507,
|
|
"cached_followup": 1008,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 1024,
|
|
"session": "0fc46cc27a844482b9eab2fef48a56d2",
|
|
"t_step1_client_unix": 1779885642.7988534,
|
|
"t_step2_client_unix": 1779885642.8733544,
|
|
"t_step2_end_unix": 1779885642.9005985,
|
|
"t_prefill_s": 0.074467666039709,
|
|
"t_transfer_s": 0.027220223972108215,
|
|
"t_followup_s": 0.0172900699544698,
|
|
"cached_followup": 1008,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 1024,
|
|
"session": "dac7cceb94c0465abce6bff73119de06",
|
|
"t_step1_client_unix": 1779885642.9183617,
|
|
"t_step2_client_unix": 1779885642.9924595,
|
|
"t_step2_end_unix": 1779885643.0189831,
|
|
"t_prefill_s": 0.07407327799592167,
|
|
"t_transfer_s": 0.026499781000893563,
|
|
"t_followup_s": 0.01622009096900001,
|
|
"cached_followup": 1008,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 1024,
|
|
"session": "4137a7ebecf148db8aa61249ef0d0100",
|
|
"t_step1_client_unix": 1779885643.0356703,
|
|
"t_step2_client_unix": 1779885643.109194,
|
|
"t_step2_end_unix": 1779885643.135519,
|
|
"t_prefill_s": 0.07349911500932649,
|
|
"t_transfer_s": 0.02630024799145758,
|
|
"t_followup_s": 0.01637468097032979,
|
|
"cached_followup": 1008,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 1024,
|
|
"session": "e28432915bf64c4891dc5157d0bebe41",
|
|
"t_step1_client_unix": 1779885643.1523507,
|
|
"t_step2_client_unix": 1779885643.2252986,
|
|
"t_step2_end_unix": 1779885643.2520146,
|
|
"t_prefill_s": 0.07292484800564125,
|
|
"t_transfer_s": 0.026693789986893535,
|
|
"t_followup_s": 0.016969401971437037,
|
|
"cached_followup": 1008,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 2048,
|
|
"session": "717a42513a2c48358ebfbb73ca07c3fd",
|
|
"t_step1_client_unix": 1779885643.269825,
|
|
"t_step2_client_unix": 1779885643.3983366,
|
|
"t_step2_end_unix": 1779885643.417481,
|
|
"t_prefill_s": 0.12848620803561062,
|
|
"t_transfer_s": 0.019112227018922567,
|
|
"t_followup_s": 0.017094306997023523,
|
|
"cached_followup": 2032,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 2048,
|
|
"session": "f2d1287f0e564a60a0d8476bb917ec9a",
|
|
"t_step1_client_unix": 1779885643.4353812,
|
|
"t_step2_client_unix": 1779885643.5640473,
|
|
"t_step2_end_unix": 1779885643.5819182,
|
|
"t_prefill_s": 0.12862860300811008,
|
|
"t_transfer_s": 0.01783871802035719,
|
|
"t_followup_s": 0.016443474043626338,
|
|
"cached_followup": 2032,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 2048,
|
|
"session": "cff5a22ae9c341d8acc455188ecc63b5",
|
|
"t_step1_client_unix": 1779885643.5992126,
|
|
"t_step2_client_unix": 1779885643.7292078,
|
|
"t_step2_end_unix": 1779885643.766638,
|
|
"t_prefill_s": 0.12995763396611437,
|
|
"t_transfer_s": 0.037401642999611795,
|
|
"t_followup_s": 0.018064863979816437,
|
|
"cached_followup": 2032,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 2048,
|
|
"session": "cc86ca16270b493fab68af1e889007af",
|
|
"t_step1_client_unix": 1779885643.7855458,
|
|
"t_step2_client_unix": 1779885643.9140093,
|
|
"t_step2_end_unix": 1779885643.9510815,
|
|
"t_prefill_s": 0.12843843002337962,
|
|
"t_transfer_s": 0.03704743104754016,
|
|
"t_followup_s": 0.017415895010344684,
|
|
"cached_followup": 2032,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 2048,
|
|
"session": "85f348460077474ab5c7f6db23005f48",
|
|
"t_step1_client_unix": 1779885643.9692879,
|
|
"t_step2_client_unix": 1779885644.0971973,
|
|
"t_step2_end_unix": 1779885644.1345108,
|
|
"t_prefill_s": 0.12788102397462353,
|
|
"t_transfer_s": 0.037287415005266666,
|
|
"t_followup_s": 0.01738263398874551,
|
|
"cached_followup": 2032,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 4096,
|
|
"session": "f715284b17af4bc29ccb689f21ac61b7",
|
|
"t_step1_client_unix": 1779885644.1534681,
|
|
"t_step2_client_unix": 1779885644.4121828,
|
|
"t_step2_end_unix": 1779885644.4760094,
|
|
"t_prefill_s": 0.25868210894986987,
|
|
"t_transfer_s": 0.06378706597024575,
|
|
"t_followup_s": 0.020318155991844833,
|
|
"cached_followup": 4080,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 4096,
|
|
"session": "b99bf61a90d54373a57af6a38d67581d",
|
|
"t_step1_client_unix": 1779885644.497922,
|
|
"t_step2_client_unix": 1779885644.7539463,
|
|
"t_step2_end_unix": 1779885644.8172574,
|
|
"t_prefill_s": 0.255985789000988,
|
|
"t_transfer_s": 0.06327042501652613,
|
|
"t_followup_s": 0.021031456999480724,
|
|
"cached_followup": 4080,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 4096,
|
|
"session": "69925f5b50f74c48800999908730d9ec",
|
|
"t_step1_client_unix": 1779885644.839903,
|
|
"t_step2_client_unix": 1779885645.0941288,
|
|
"t_step2_end_unix": 1779885645.1562943,
|
|
"t_prefill_s": 0.2541897820192389,
|
|
"t_transfer_s": 0.06213490100344643,
|
|
"t_followup_s": 0.021715508948545903,
|
|
"cached_followup": 4080,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 4096,
|
|
"session": "312743112bc84c0699a91e3110a63f03",
|
|
"t_step1_client_unix": 1779885645.1795704,
|
|
"t_step2_client_unix": 1779885645.4336376,
|
|
"t_step2_end_unix": 1779885645.4955726,
|
|
"t_prefill_s": 0.2540306959999725,
|
|
"t_transfer_s": 0.06190510099986568,
|
|
"t_followup_s": 0.020819986995775253,
|
|
"cached_followup": 4080,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 4096,
|
|
"session": "e5975bd25e1b4aa586cbfe9b68370de2",
|
|
"t_step1_client_unix": 1779885645.5179617,
|
|
"t_step2_client_unix": 1779885645.7724028,
|
|
"t_step2_end_unix": 1779885645.8355522,
|
|
"t_prefill_s": 0.2544059020001441,
|
|
"t_transfer_s": 0.06311210000421852,
|
|
"t_followup_s": 0.02235932502662763,
|
|
"cached_followup": 4080,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 8192,
|
|
"session": "f8989efa9261470e81c2891945b460a5",
|
|
"t_step1_client_unix": 1779885645.8609917,
|
|
"t_step2_client_unix": 1779885646.4364054,
|
|
"t_step2_end_unix": 1779885646.4636223,
|
|
"t_prefill_s": 0.575372110994067,
|
|
"t_transfer_s": 0.02718952501891181,
|
|
"t_followup_s": 0.024976509041152894,
|
|
"cached_followup": 8176,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 8192,
|
|
"session": "3a8632e8562c4cbcaa32b57ba23bbf68",
|
|
"t_step1_client_unix": 1779885646.4916308,
|
|
"t_step2_client_unix": 1779885647.0670328,
|
|
"t_step2_end_unix": 1779885647.0934644,
|
|
"t_prefill_s": 0.5753633370040916,
|
|
"t_transfer_s": 0.026399013004265726,
|
|
"t_followup_s": 0.027285586984362453,
|
|
"cached_followup": 8176,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 8192,
|
|
"session": "4d2cfc02defd492d89e90f652032f630",
|
|
"t_step1_client_unix": 1779885647.1237462,
|
|
"t_step2_client_unix": 1779885647.6985745,
|
|
"t_step2_end_unix": 1779885647.8096595,
|
|
"t_prefill_s": 0.574782071984373,
|
|
"t_transfer_s": 0.11104170099133626,
|
|
"t_followup_s": 0.02618999598780647,
|
|
"cached_followup": 8176,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 8192,
|
|
"session": "3485a775304947ada42d61cfd39b7520",
|
|
"t_step1_client_unix": 1779885647.8389955,
|
|
"t_step2_client_unix": 1779885648.4133987,
|
|
"t_step2_end_unix": 1779885648.5235367,
|
|
"t_prefill_s": 0.574362020008266,
|
|
"t_transfer_s": 0.11010575300315395,
|
|
"t_followup_s": 0.028998883964959532,
|
|
"cached_followup": 8176,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 8192,
|
|
"session": "6f87692fb5884094b7b141116ba739f8",
|
|
"t_step1_client_unix": 1779885648.5556417,
|
|
"t_step2_client_unix": 1779885649.1294918,
|
|
"t_step2_end_unix": 1779885649.2386518,
|
|
"t_prefill_s": 0.5738097460125573,
|
|
"t_transfer_s": 0.1091307660099119,
|
|
"t_followup_s": 0.025612210971303284,
|
|
"cached_followup": 8176,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 16384,
|
|
"session": "c4e007ebdcf244cf95a8ccb3ab59ea0f",
|
|
"t_step1_client_unix": 1779885649.270192,
|
|
"t_step2_client_unix": 1779885650.7972357,
|
|
"t_step2_end_unix": 1779885651.0219116,
|
|
"t_prefill_s": 1.5270042870542966,
|
|
"t_transfer_s": 0.22463428904302418,
|
|
"t_followup_s": 0.03729638300137594,
|
|
"cached_followup": 16368,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 16384,
|
|
"session": "6f34ab6672e14f41b7780323b30ce21e",
|
|
"t_step1_client_unix": 1779885651.0651574,
|
|
"t_step2_client_unix": 1779885652.5826244,
|
|
"t_step2_end_unix": 1779885652.7871268,
|
|
"t_prefill_s": 1.5174251759890467,
|
|
"t_transfer_s": 0.2044643560075201,
|
|
"t_followup_s": 0.03547715302556753,
|
|
"cached_followup": 16368,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 16384,
|
|
"session": "ca02bd5aa32544f2b1b09096e5f63c8b",
|
|
"t_step1_client_unix": 1779885652.8284712,
|
|
"t_step2_client_unix": 1779885654.3465412,
|
|
"t_step2_end_unix": 1779885654.5522587,
|
|
"t_prefill_s": 1.5180304029490799,
|
|
"t_transfer_s": 0.20567574101733044,
|
|
"t_followup_s": 0.03611186001216993,
|
|
"cached_followup": 16368,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 16384,
|
|
"session": "59047ab966574f72ab12693ad29b090f",
|
|
"t_step1_client_unix": 1779885654.5942106,
|
|
"t_step2_client_unix": 1779885656.1123874,
|
|
"t_step2_end_unix": 1779885656.3158703,
|
|
"t_prefill_s": 1.5181243289844133,
|
|
"t_transfer_s": 0.20344976399792358,
|
|
"t_followup_s": 0.035486461012624204,
|
|
"cached_followup": 16368,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 16384,
|
|
"session": "9a6e2228fd40414c8fd2b78bf399552c",
|
|
"t_step1_client_unix": 1779885656.357184,
|
|
"t_step2_client_unix": 1779885657.8739264,
|
|
"t_step2_end_unix": 1779885658.079835,
|
|
"t_prefill_s": 1.5166968539706431,
|
|
"t_transfer_s": 0.20585927896900102,
|
|
"t_followup_s": 0.03941280301660299,
|
|
"cached_followup": 16368,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 32768,
|
|
"session": "5cf9fb856b674c0a9237c06980287cb3",
|
|
"t_step1_client_unix": 1779885658.1312225,
|
|
"t_step2_client_unix": 1779885662.6780548,
|
|
"t_step2_end_unix": 1779885663.0500648,
|
|
"t_prefill_s": 4.54679255298106,
|
|
"t_transfer_s": 0.3719712099991739,
|
|
"t_followup_s": 0.05820785299874842,
|
|
"cached_followup": 32752,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 32768,
|
|
"session": "524f192c92834453bc48052bc6516126",
|
|
"t_step1_client_unix": 1779885663.1199641,
|
|
"t_step2_client_unix": 1779885667.6679316,
|
|
"t_step2_end_unix": 1779885668.047607,
|
|
"t_prefill_s": 4.547927976993378,
|
|
"t_transfer_s": 0.3796397229889408,
|
|
"t_followup_s": 0.058314190013334155,
|
|
"cached_followup": 32752,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 32768,
|
|
"session": "fb7aa2c99bf24febb83536c54e30a36d",
|
|
"t_step1_client_unix": 1779885668.117419,
|
|
"t_step2_client_unix": 1779885672.6644971,
|
|
"t_step2_end_unix": 1779885673.0345452,
|
|
"t_prefill_s": 4.547034470015205,
|
|
"t_transfer_s": 0.3700092720100656,
|
|
"t_followup_s": 0.05658108199713752,
|
|
"cached_followup": 32752,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 32768,
|
|
"session": "85612444c3224b65afb09d65a829b5cc",
|
|
"t_step1_client_unix": 1779885673.1025856,
|
|
"t_step2_client_unix": 1779885677.6487892,
|
|
"t_step2_end_unix": 1779885678.2856307,
|
|
"t_prefill_s": 4.546165039995685,
|
|
"t_transfer_s": 0.6368027949938551,
|
|
"t_followup_s": 0.05732736398931593,
|
|
"cached_followup": 32752,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 32768,
|
|
"session": "fe51fbacc4d544b3a500b099f852b19b",
|
|
"t_step1_client_unix": 1779885678.3544915,
|
|
"t_step2_client_unix": 1779885682.9011996,
|
|
"t_step2_end_unix": 1779885683.542081,
|
|
"t_prefill_s": 4.546669405011926,
|
|
"t_transfer_s": 0.6408427829737775,
|
|
"t_followup_s": 0.057976288022473454,
|
|
"cached_followup": 32752,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 65536,
|
|
"session": "9628b04d4d9e4a9fb26decccef84d32f",
|
|
"t_step1_client_unix": 1779885683.6233714,
|
|
"t_step2_client_unix": 1779885698.8039112,
|
|
"t_step2_end_unix": 1779885700.9075115,
|
|
"t_prefill_s": 15.18050062697148,
|
|
"t_transfer_s": 2.103562032978516,
|
|
"t_followup_s": 0.0921451430185698,
|
|
"cached_followup": 65520,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 65536,
|
|
"session": "b8bd7328d2274613b473e932db11081c",
|
|
"t_step1_client_unix": 1779885701.0228417,
|
|
"t_step2_client_unix": 1779885716.2263126,
|
|
"t_step2_end_unix": 1779885718.3809934,
|
|
"t_prefill_s": 15.200471161981113,
|
|
"t_transfer_s": 2.1520405350020155,
|
|
"t_followup_s": 0.11590708000585437,
|
|
"cached_followup": 65520,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 65536,
|
|
"session": "8ca1dfe708c945e59c194ee59b7de87d",
|
|
"t_step1_client_unix": 1779885718.5226529,
|
|
"t_step2_client_unix": 1779885733.706992,
|
|
"t_step2_end_unix": 1779885735.6409419,
|
|
"t_prefill_s": 15.184306180977728,
|
|
"t_transfer_s": 1.9339170460007153,
|
|
"t_followup_s": 0.09697375196265057,
|
|
"cached_followup": 65520,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 65536,
|
|
"session": "aa7555475c87427ab1a1d95277545655",
|
|
"t_step1_client_unix": 1779885735.7609437,
|
|
"t_step2_client_unix": 1779885750.9334345,
|
|
"t_step2_end_unix": 1779885752.0260077,
|
|
"t_prefill_s": 15.172441756993067,
|
|
"t_transfer_s": 1.09253996796906,
|
|
"t_followup_s": 0.08802600798662752,
|
|
"cached_followup": 65520,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 65536,
|
|
"session": "dece39eeadd645c7a8e563763f93c35f",
|
|
"t_step1_client_unix": 1779885752.1370633,
|
|
"t_step2_client_unix": 1779885767.3073487,
|
|
"t_step2_end_unix": 1779885768.0632908,
|
|
"t_prefill_s": 15.17024433100596,
|
|
"t_transfer_s": 0.7559080219944008,
|
|
"t_followup_s": 0.08695380098652095,
|
|
"cached_followup": 65520,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 131072,
|
|
"session": "98a339dad05a463e8ac1425413221d27",
|
|
"t_step1_client_unix": 1779885768.196416,
|
|
"t_step2_client_unix": 1779885822.9193504,
|
|
"t_step2_end_unix": 1779885824.3877244,
|
|
"t_prefill_s": 54.72287795698503,
|
|
"t_transfer_s": 1.4683381259674206,
|
|
"t_followup_s": 0.16380483901593834,
|
|
"cached_followup": 131056,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 131072,
|
|
"session": "4ce2151ad3d4493397bb77c7a62871e7",
|
|
"t_step1_client_unix": 1779885824.5981278,
|
|
"t_step2_client_unix": 1779885879.3268123,
|
|
"t_step2_end_unix": 1779885884.7123976,
|
|
"t_prefill_s": 54.72864995297277,
|
|
"t_transfer_s": 5.38555136800278,
|
|
"t_followup_s": 0.17249851900851354,
|
|
"cached_followup": 131056,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 131072,
|
|
"session": "6a214438dcf74f36bbefaaed3583c04c",
|
|
"t_step1_client_unix": 1779885884.9482107,
|
|
"t_step2_client_unix": 1779885939.6747935,
|
|
"t_step2_end_unix": 1779885949.2077906,
|
|
"t_prefill_s": 54.72654917498585,
|
|
"t_transfer_s": 9.532963149016723,
|
|
"t_followup_s": 0.1658564520184882,
|
|
"cached_followup": 131056,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 131072,
|
|
"session": "4cd33910256e43768a2f475650d7714c",
|
|
"t_step1_client_unix": 1779885949.4215908,
|
|
"t_step2_client_unix": 1779886004.145101,
|
|
"t_step2_end_unix": 1779886007.2428164,
|
|
"t_prefill_s": 54.72347703698324,
|
|
"t_transfer_s": 3.0976800689822994,
|
|
"t_followup_s": 0.16629640298197046,
|
|
"cached_followup": 131056,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
},
|
|
{
|
|
"input_tokens": 131072,
|
|
"session": "7e3626253b5d4fe29b3cec1e3598bb39",
|
|
"t_step1_client_unix": 1779886007.4549844,
|
|
"t_step2_client_unix": 1779886062.0564816,
|
|
"t_step2_end_unix": 1779886063.7279255,
|
|
"t_prefill_s": 54.60145856201416,
|
|
"t_transfer_s": 1.6714086790452711,
|
|
"t_followup_s": 0.1628041280200705,
|
|
"cached_followup": 131056,
|
|
"pull_completion_tokens": 1,
|
|
"ok": true
|
|
}
|
|
],
|
|
"summary": [
|
|
{
|
|
"input_tokens": 512,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 0.019290994200855494,
|
|
"transfer_s_p50": 0.019964047998655587,
|
|
"transfer_s_p90": 0.020512140006758273,
|
|
"transfer_s_min": 0.016899197013117373,
|
|
"transfer_s_max": 0.020512140006758273
|
|
},
|
|
{
|
|
"input_tokens": 1024,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 0.027316297392826528,
|
|
"transfer_s_p50": 0.026693789986893535,
|
|
"transfer_s_p90": 0.029867444012779742,
|
|
"transfer_s_min": 0.02630024799145758,
|
|
"transfer_s_max": 0.029867444012779742
|
|
},
|
|
{
|
|
"input_tokens": 2048,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 0.029737486818339674,
|
|
"transfer_s_p50": 0.03704743104754016,
|
|
"transfer_s_p90": 0.037401642999611795,
|
|
"transfer_s_min": 0.01783871802035719,
|
|
"transfer_s_max": 0.037401642999611795
|
|
},
|
|
{
|
|
"input_tokens": 4096,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 0.0628419185988605,
|
|
"transfer_s_p50": 0.06311210000421852,
|
|
"transfer_s_p90": 0.06378706597024575,
|
|
"transfer_s_min": 0.06190510099986568,
|
|
"transfer_s_max": 0.06378706597024575
|
|
},
|
|
{
|
|
"input_tokens": 8192,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 0.07677335160551593,
|
|
"transfer_s_p50": 0.1091307660099119,
|
|
"transfer_s_p90": 0.11104170099133626,
|
|
"transfer_s_min": 0.026399013004265726,
|
|
"transfer_s_max": 0.11104170099133626
|
|
},
|
|
{
|
|
"input_tokens": 16384,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 0.20881668580695986,
|
|
"transfer_s_p50": 0.20567574101733044,
|
|
"transfer_s_p90": 0.22463428904302418,
|
|
"transfer_s_min": 0.20344976399792358,
|
|
"transfer_s_max": 0.22463428904302418
|
|
},
|
|
{
|
|
"input_tokens": 32768,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 0.4798531565931626,
|
|
"transfer_s_p50": 0.3796397229889408,
|
|
"transfer_s_p90": 0.6408427829737775,
|
|
"transfer_s_min": 0.3700092720100656,
|
|
"transfer_s_max": 0.6408427829737775
|
|
},
|
|
{
|
|
"input_tokens": 65536,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 1.6075935207889416,
|
|
"transfer_s_p50": 1.9339170460007153,
|
|
"transfer_s_p90": 2.1520405350020155,
|
|
"transfer_s_min": 0.7559080219944008,
|
|
"transfer_s_max": 2.1520405350020155
|
|
},
|
|
{
|
|
"input_tokens": 131072,
|
|
"n_ok": 5,
|
|
"transfer_s_mean": 4.2311882782028984,
|
|
"transfer_s_p50": 3.0976800689822994,
|
|
"transfer_s_p90": 9.532963149016723,
|
|
"transfer_s_min": 1.4683381259674206,
|
|
"transfer_s_max": 9.532963149016723
|
|
}
|
|
]
|
|
} |