mudassiruddin21 commited on
Commit
e302ab2
·
verified ·
1 Parent(s): 38db027

Run r3: logs and results

Browse files
runs/r3/engine_logs/20261119-000859-eval-r1.log CHANGED
@@ -244,3 +244,42 @@ INFO: 127.0.0.1:31314 - "GET /metrics HTTP/1.1" 200 OK
244
  INFO: 127.0.0.1:19036 - "GET /metrics HTTP/1.1" 200 OK
245
  INFO: 127.0.0.1:51504 - "POST /v1/completions HTTP/1.1" 200 OK
246
  INFO: 127.0.0.1:50818 - "POST /v1/completions HTTP/1.1" 200 OK
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
244
  INFO: 127.0.0.1:19036 - "GET /metrics HTTP/1.1" 200 OK
245
  INFO: 127.0.0.1:51504 - "POST /v1/completions HTTP/1.1" 200 OK
246
  INFO: 127.0.0.1:50818 - "POST /v1/completions HTTP/1.1" 200 OK
247
+ INFO: 127.0.0.1:17249 - "GET /metrics HTTP/1.1" 200 OK
248
+ INFO: 127.0.0.1:47130 - "GET /metrics HTTP/1.1" 200 OK
249
+ INFO: 127.0.0.1:63847 - "POST /v1/completions HTTP/1.1" 200 OK
250
+ (APIServer pid=14373) INFO 11-19 00:14:52 [loggers.py:320] Engine 000: Avg prompt throughput: 39.2 tokens/s, Avg generation throughput: 66.8 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
251
+ (APIServer pid=14373) INFO 11-19 00:14:52 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.75, Accepted throughput: 48.90 tokens/s, Drafted throughput: 53.40 tokens/s, Accepted: 489 tokens, Drafted: 534 tokens, Per-position acceptance rate: 0.961, 0.921, 0.865, Avg Draft acceptance rate: 91.6%
252
+ INFO: 127.0.0.1:38622 - "POST /v1/completions HTTP/1.1" 200 OK
253
+ (APIServer pid=14373) INFO 11-19 00:15:02 [loggers.py:320] Engine 000: Avg prompt throughput: 12.1 tokens/s, Avg generation throughput: 67.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
254
+ (APIServer pid=14373) INFO 11-19 00:15:02 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 3.69, Accepted throughput: 49.19 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 492 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.967, 0.902, 0.820, Avg Draft acceptance rate: 89.6%
255
+ INFO: 127.0.0.1:21154 - "GET /metrics HTTP/1.1" 200 OK
256
+ INFO: 127.0.0.1:64017 - "GET /metrics HTTP/1.1" 200 OK
257
+ INFO: 127.0.0.1:62658 - "POST /v1/completions HTTP/1.1" 200 OK
258
+ (APIServer pid=14373) INFO 11-19 00:15:12 [loggers.py:320] Engine 000: Avg prompt throughput: 4.2 tokens/s, Avg generation throughput: 44.7 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
259
+ (APIServer pid=14373) INFO 11-19 00:15:12 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.44, Accepted throughput: 26.30 tokens/s, Drafted throughput: 54.89 tokens/s, Accepted: 263 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.721, 0.443, 0.273, Avg Draft acceptance rate: 47.9%
260
+ (APIServer pid=14373) INFO 11-19 00:15:22 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 45.0 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.4%
261
+ (APIServer pid=14373) INFO 11-19 00:15:22 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.43, Accepted throughput: 26.50 tokens/s, Drafted throughput: 55.49 tokens/s, Accepted: 265 tokens, Drafted: 555 tokens, Per-position acceptance rate: 0.784, 0.411, 0.238, Avg Draft acceptance rate: 47.7%
262
+ INFO: 127.0.0.1:61318 - "POST /v1/completions HTTP/1.1" 200 OK
263
+ (APIServer pid=14373) INFO 11-19 00:15:32 [loggers.py:320] Engine 000: Avg prompt throughput: 3.8 tokens/s, Avg generation throughput: 41.4 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
264
+ (APIServer pid=14373) INFO 11-19 00:15:32 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.26, Accepted throughput: 23.10 tokens/s, Drafted throughput: 54.90 tokens/s, Accepted: 231 tokens, Drafted: 549 tokens, Per-position acceptance rate: 0.672, 0.372, 0.219, Avg Draft acceptance rate: 42.1%
265
+ (APIServer pid=14373) INFO 11-19 00:15:42 [loggers.py:320] Engine 000: Avg prompt throughput: 0.0 tokens/s, Avg generation throughput: 40.6 tokens/s, Running: 1 reqs, Waiting: 0 reqs, GPU KV cache usage: 2.2%
266
+ (APIServer pid=14373) INFO 11-19 00:15:42 [metrics.py:120] SpecDecoding metrics: Mean acceptance length: 2.21, Accepted throughput: 22.20 tokens/s, Drafted throughput: 55.19 tokens/s, Accepted: 222 tokens, Drafted: 552 tokens, Per-position acceptance rate: 0.696, 0.353, 0.158, Avg Draft acceptance rate: 40.2%
267
+ INFO: 127.0.0.1:53207 - "GET /metrics HTTP/1.1" 200 OK
268
+ (APIServer pid=14373) INFO 11-19 00:15:44 [launcher.py:128] [shutdown] API server: shutdown triggered
269
+ (APIServer pid=14373) INFO 11-19 00:15:44 [launcher.py:144] [shutdown] API server: stopping engine client mode=abort timeout=0s
270
+ (APIServer pid=14373) INFO 11-19 00:15:44 [core_client.py:762] [shutdown] MPClient: start timeout=0s
271
+ (APIServer pid=14373) INFO 11-19 00:15:44 [core_client.py:764] [shutdown] MPClient: stopping engine manager
272
+ (APIServer pid=14373) INFO 11-19 00:15:44 [utils.py:638] [shutdown] Process manager: send sigterm to process EngineCore
273
+ (APIServer pid=14373) WARNING 11-19 00:15:44 [utils.py:658] [shutdown] Process manager: force killing remaining processes count=1
274
+ (APIServer pid=14373) WARNING 11-19 00:15:44 [utils.py:663] [shutdown] Process manager: force killing remaining process EngineCore pid 14912
275
+ (APIServer pid=14373) INFO 11-19 00:15:44 [core_client.py:766] [shutdown] MPClient: engine manager stopped
276
+ (APIServer pid=14373) INFO 11-19 00:15:44 [core_client.py:767] [shutdown] MPClient: cleaning up background resources
277
+ (APIServer pid=14373) INFO 11-19 00:15:44 [core_client.py:769] [shutdown] MPClient: complete
278
+ (APIServer pid=14373) INFO 11-19 00:15:44 [launcher.py:153] [shutdown] API server: engine client stopped
279
+ (APIServer pid=14373) INFO 11-19 00:15:44 [launcher.py:156] [shutdown] API server: signalling HTTP server shutdown
280
+ (APIServer pid=14373) INFO 11-19 00:15:44 [launcher.py:178] [shutdown] API server: shutting down FastAPI HTTP server
281
+ INFO: Shutting down
282
+ INFO: Waiting for application shutdown.
283
+ INFO: Application shutdown complete.
284
+ /workspace/emc/uv-python/cpython-3.12.13-linux-x86_64-gnu/lib/python3.12/multiprocessing/resource_tracker.py:279: UserWarning: resource_tracker: There appear to be 1 leaked semaphore objects to clean up at shutdown
285
+ warnings.warn('resource_tracker: There appear to be %d '
runs/r3/kit_results/20261119-001545-results.json ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "mtp",
3
+ "prompts": 8,
4
+ "repeats": 1,
5
+ "acceptance_length": 2.8386876281613125,
6
+ "acceptance_by_domain": {
7
+ "chat": 3.0507462686567166,
8
+ "code": 3.7413793103448274,
9
+ "math": 3.7010309278350517,
10
+ "longform": 2.318421052631579
11
+ },
12
+ "tpot_median_s": 0.014886422647702478,
13
+ "ttft_median_s": 0.11165462700000717,
14
+ "memory_bytes": 607177728,
15
+ "weight_bytes": 540068864,
16
+ "kv_cache_bytes": 67108864,
17
+ "decode_total_s": 80.02569263999999,
18
+ "num_speculative_tokens": 3,
19
+ "max_model_len": 16384,
20
+ "engine": {
21
+ "vllm_version": "0.31.0",
22
+ "vllm_version_pinned": "0.31.0",
23
+ "image": "pip:vllm==0.31.0",
24
+ "matches_pin": true
25
+ },
26
+ "per_prompt": [
27
+ {
28
+ "id": "chat-0000",
29
+ "domain": "chat",
30
+ "tokens": 512,
31
+ "ttft_s": 0.10816,
32
+ "total_s": 10.53706,
33
+ "tpot_s": 0.020409
34
+ },
35
+ {
36
+ "id": "chat-0001",
37
+ "domain": "chat",
38
+ "tokens": 512,
39
+ "ttft_s": 0.11157,
40
+ "total_s": 7.78227,
41
+ "tpot_s": 0.015011
42
+ },
43
+ {
44
+ "id": "code-0000",
45
+ "domain": "code",
46
+ "tokens": 139,
47
+ "ttft_s": 0.10677,
48
+ "total_s": 1.99512,
49
+ "tpot_s": 0.013684
50
+ },
51
+ {
52
+ "id": "code-0001",
53
+ "domain": "code",
54
+ "tokens": 512,
55
+ "ttft_s": 0.12178,
56
+ "total_s": 7.64619,
57
+ "tpot_s": 0.014725
58
+ },
59
+ {
60
+ "id": "math-0000",
61
+ "domain": "math",
62
+ "tokens": 384,
63
+ "ttft_s": 0.10432,
64
+ "total_s": 5.66715,
65
+ "tpot_s": 0.014524
66
+ },
67
+ {
68
+ "id": "math-0001",
69
+ "domain": "math",
70
+ "tokens": 334,
71
+ "ttft_s": 0.1159,
72
+ "total_s": 5.03155,
73
+ "tpot_s": 0.014762
74
+ },
75
+ {
76
+ "id": "longform-0000",
77
+ "domain": "longform",
78
+ "tokens": 933,
79
+ "ttft_s": 0.11741,
80
+ "total_s": 21.23246,
81
+ "tpot_s": 0.022656
82
+ },
83
+ {
84
+ "id": "longform-0001",
85
+ "domain": "longform",
86
+ "tokens": 827,
87
+ "ttft_s": 0.11173,
88
+ "total_s": 20.13389,
89
+ "tpot_s": 0.02424
90
+ }
91
+ ],
92
+ "score": {
93
+ "score": 110.38,
94
+ "ranked": true,
95
+ "s_acceptance": 1.0201,
96
+ "s_memory": 1.5,
97
+ "s_tpot": 0.8749
98
+ }
99
+ }
runs/r3/logs/pipeline.log CHANGED
@@ -12,3 +12,4 @@
12
  [2026-11-19 00:08:21] done probe (24s)
13
  [2026-11-19 00:08:42] done prompts (56s)
14
  [2026-11-19 00:08:51] start smoke_fp8
 
 
12
  [2026-11-19 00:08:21] done probe (24s)
13
  [2026-11-19 00:08:42] done prompts (56s)
14
  [2026-11-19 00:08:51] start smoke_fp8
15
+ [2026-11-19 00:15:46] done smoke_fp8 (415s)
runs/r3/logs/pipeline.out CHANGED
@@ -12,3 +12,4 @@
12
  [2026-11-19 00:08:21] done probe (24s)
13
  [2026-11-19 00:08:42] done prompts (56s)
14
  [2026-11-19 00:08:51] start smoke_fp8
 
 
12
  [2026-11-19 00:08:21] done probe (24s)
13
  [2026-11-19 00:08:42] done prompts (56s)
14
  [2026-11-19 00:08:51] start smoke_fp8
15
+ [2026-11-19 00:15:46] done smoke_fp8 (415s)
runs/r3/logs/smoke_fp8.log CHANGED
@@ -8,3 +8,25 @@ PASS
8
  [00:08:59] starting vLLM (log: /workspace/emc/work/logs/20261119-000859-eval-r1.log)
9
  [00:13:13] vLLM ready after 254s
10
  [00:14:42] chat: 2 prompts done
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  [00:08:59] starting vLLM (log: /workspace/emc/work/logs/20261119-000859-eval-r1.log)
9
  [00:13:13] vLLM ready after 254s
10
  [00:14:42] chat: 2 prompts done
11
+ [00:14:52] code: 2 prompts done
12
+ [00:15:03] math: 2 prompts done
13
+ [00:15:44] longform: 2 prompts done
14
+
15
+ ================ emc2 eval summary ================
16
+ engine vLLM 0.31.0
17
+ method mtp (K = 3, context 16384)
18
+ prompts x repeats 8 x 1
19
+ acceptance length 2.839
20
+ chat 3.051
21
+ code 3.741
22
+ math 3.701
23
+ longform 2.318
24
+ TPOT (median) 14.89 ms per output token (this GPU)
25
+ TTFT (median, info) 111.7 ms (this GPU; not scored)
26
+ memory metric 0.607 GB (weights 0.540 + KV 0.067)
27
+ score 110.4 (baseline = 100)
28
+ components acceptance 1.020, memory 1.500, tpot 0.875
29
+ full results /workspace/emc/work/results/20261119-001545-results.json
30
+ ===================================================
31
+
32
+ FP8 probe acceptance 2.839 vs published baseline 2.802 (ratio 1.020)
runs/r3/logs/stats.log CHANGED
@@ -11,3 +11,4 @@
11
  [2026-11-19 00:08:42] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.1 GB
12
  [2026-11-19 00:08:51] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.1 GB
13
  [2026-11-19 00:14:47] ram 5.0/172 GB used (cache 0.3, shmem 0.0) | cgroup 5.0/172 GB | disk 66 GB used | gpu 83.1/95.6 GiB, 100%, 288 W | load 0.0 | largest VLLM::EngineCor 4.7 GB, vllm 2.5 GB, python 1.2 GB
 
 
11
  [2026-11-19 00:08:42] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.1 GB
12
  [2026-11-19 00:08:51] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 59 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.1 GB
13
  [2026-11-19 00:14:47] ram 5.0/172 GB used (cache 0.3, shmem 0.0) | cgroup 5.0/172 GB | disk 66 GB used | gpu 83.1/95.6 GiB, 100%, 288 W | load 0.0 | largest VLLM::EngineCor 4.7 GB, vllm 2.5 GB, python 1.2 GB
14
+ [2026-11-19 00:15:46] ram 0.2/172 GB used (cache 0.0, shmem 0.0) | cgroup 0.2/172 GB | disk 66 GB used | gpu 0.0/95.6 GiB, 0%, 85 W | load 0.0 | largest python 0.1 GB, python 0.1 GB, python 0.0 GB
runs/r3/logs/sync.log CHANGED
@@ -4,3 +4,4 @@ uploaded logs/results to mudassiruddin21/EMC:runs/r3
4
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
5
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
6
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
 
 
4
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
5
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
6
  uploaded logs/results to mudassiruddin21/EMC:runs/r3
7
+ uploaded logs/results to mudassiruddin21/EMC:runs/r3
runs/r3/results/smoke_fp8.json ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "method": "mtp",
3
+ "prompts": 8,
4
+ "repeats": 1,
5
+ "acceptance_length": 2.8386876281613125,
6
+ "acceptance_by_domain": {
7
+ "chat": 3.0507462686567166,
8
+ "code": 3.7413793103448274,
9
+ "math": 3.7010309278350517,
10
+ "longform": 2.318421052631579
11
+ },
12
+ "tpot_median_s": 0.014886422647702478,
13
+ "ttft_median_s": 0.11165462700000717,
14
+ "memory_bytes": 607177728,
15
+ "weight_bytes": 540068864,
16
+ "kv_cache_bytes": 67108864,
17
+ "decode_total_s": 80.02569263999999,
18
+ "num_speculative_tokens": 3,
19
+ "max_model_len": 16384,
20
+ "engine": {
21
+ "vllm_version": "0.31.0",
22
+ "vllm_version_pinned": "0.31.0",
23
+ "image": "pip:vllm==0.31.0",
24
+ "matches_pin": true
25
+ },
26
+ "per_prompt": [
27
+ {
28
+ "id": "chat-0000",
29
+ "domain": "chat",
30
+ "tokens": 512,
31
+ "ttft_s": 0.10816,
32
+ "total_s": 10.53706,
33
+ "tpot_s": 0.020409
34
+ },
35
+ {
36
+ "id": "chat-0001",
37
+ "domain": "chat",
38
+ "tokens": 512,
39
+ "ttft_s": 0.11157,
40
+ "total_s": 7.78227,
41
+ "tpot_s": 0.015011
42
+ },
43
+ {
44
+ "id": "code-0000",
45
+ "domain": "code",
46
+ "tokens": 139,
47
+ "ttft_s": 0.10677,
48
+ "total_s": 1.99512,
49
+ "tpot_s": 0.013684
50
+ },
51
+ {
52
+ "id": "code-0001",
53
+ "domain": "code",
54
+ "tokens": 512,
55
+ "ttft_s": 0.12178,
56
+ "total_s": 7.64619,
57
+ "tpot_s": 0.014725
58
+ },
59
+ {
60
+ "id": "math-0000",
61
+ "domain": "math",
62
+ "tokens": 384,
63
+ "ttft_s": 0.10432,
64
+ "total_s": 5.66715,
65
+ "tpot_s": 0.014524
66
+ },
67
+ {
68
+ "id": "math-0001",
69
+ "domain": "math",
70
+ "tokens": 334,
71
+ "ttft_s": 0.1159,
72
+ "total_s": 5.03155,
73
+ "tpot_s": 0.014762
74
+ },
75
+ {
76
+ "id": "longform-0000",
77
+ "domain": "longform",
78
+ "tokens": 933,
79
+ "ttft_s": 0.11741,
80
+ "total_s": 21.23246,
81
+ "tpot_s": 0.022656
82
+ },
83
+ {
84
+ "id": "longform-0001",
85
+ "domain": "longform",
86
+ "tokens": 827,
87
+ "ttft_s": 0.11173,
88
+ "total_s": 20.13389,
89
+ "tpot_s": 0.02424
90
+ }
91
+ ],
92
+ "score": {
93
+ "score": 110.38,
94
+ "ranked": true,
95
+ "s_acceptance": 1.0201,
96
+ "s_memory": 1.5,
97
+ "s_tpot": 0.8749
98
+ }
99
+ }
runs/r3/state/fp8_ok ADDED
@@ -0,0 +1 @@
 
 
1
+ 1.0201
runs/r3/state/smoke_fp8.done ADDED
@@ -0,0 +1 @@
 
 
1
+ 415