-
Notifications
You must be signed in to change notification settings - Fork 0
/
benchmark_temperature_tp4_hot_gpus.txt
847 lines (815 loc) · 35.8 KB
/
benchmark_temperature_tp4_hot_gpus.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Open ttft file
Started DCGMI
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.12218437592188518
P50 ttft = 0.10786616802215576
P99 ttft = 0.20806359052658083
Average tbt = 0.03983792265256246
P50 tbt = 0.02372201681137085
P99 tbt = 0.13479589581489565
All GPUs and memories are cold after 1.0234026908874512
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.11637558255876813
P50 ttft = 0.10694503784179688
P99 ttft = 0.2017923355102539
Average tbt = 0.040876186461675734
P50 tbt = 0.023461532592773438
P99 tbt = 0.15499648571014413
All GPUs and memories are cold after 6.015701770782471
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.13451288087027413
P50 ttft = 0.10668706893920898
P99 ttft = 0.3615493535995481
Average tbt = 0.06527677740369527
P50 tbt = 0.024234724044799805
P99 tbt = 0.2157229423522949
All GPUs and memories are cold after 14.025269746780396
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.12842206838654308
P50 ttft = 0.10675382614135742
P99 ttft = 0.2539934158325196
Average tbt = 0.06570955136927163
P50 tbt = 0.024153947830200195
P99 tbt = 0.28553801536560075
All GPUs and memories are cold after 16.048485279083252
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1352384853363037
P50 ttft = 0.10742485523223877
P99 ttft = 0.25277865409851064
Average tbt = 0.07749011135101318
P50 tbt = 0.02524203062057495
P99 tbt = 0.283778379201889
All GPUs and memories are cold after 20.055327653884888
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.14023305103182793
P50 ttft = 0.1106119155883789
P99 ttft = 0.27456366777420044
Average tbt = 0.09223630800843242
P50 tbt = 0.09531164169311525
P99 tbt = 0.3419849753379821
All GPUs and memories are cold after 22.141963958740234
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.14689406956711862
P50 ttft = 0.11544227600097656
P99 ttft = 0.2741622829437256
Average tbt = 0.11429604732826965
P50 tbt = 0.09589540958404542
P99 tbt = 0.45010433864593524
All GPUs and memories are cold after 25.08884620666504
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1583217109542295
P50 ttft = 0.12288594245910645
P99 ttft = 0.4082523059844969
Average tbt = 0.12449614082474308
P50 tbt = 0.09667992591857912
P99 tbt = 0.4850542984008785
All GPUs and memories are cold after 27.055988788604736
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.11607742309570312
P50 ttft = 0.10750055313110352
P99 ttft = 0.1935513091087342
Average tbt = 0.03818810780843098
P50 tbt = 0.02340489625930786
P99 tbt = 0.12456525468826296
All GPUs and memories are cold after 3.0131895542144775
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.11563659849621001
P50 ttft = 0.1062769889831543
P99 ttft = 0.20128812789916994
Average tbt = 0.040518737974621
P50 tbt = 0.023278188705444337
P99 tbt = 0.15416809558868416
All GPUs and memories are cold after 7.0574023723602295
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1254589489528111
P50 ttft = 0.10652875900268555
P99 ttft = 0.25059515476226785
Average tbt = 0.05786213125501362
P50 tbt = 0.023859691619873048
P99 tbt = 0.1894655055999755
All GPUs and memories are cold after 14.028018712997437
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.12930161197011064
P50 ttft = 0.10743093490600586
P99 ttft = 0.2531076431274415
Average tbt = 0.06562419809946199
P50 tbt = 0.023950815200805664
P99 tbt = 0.2864652872085573
All GPUs and memories are cold after 14.082073450088501
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.13547237873077392
P50 ttft = 0.1069566011428833
P99 ttft = 0.25389883279800407
Average tbt = 0.07741855955123902
P50 tbt = 0.024684214591979982
P99 tbt = 0.2845805091857909
All GPUs and memories are cold after 20.035526752471924
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.14047792926430702
P50 ttft = 0.11295056343078613
P99 ttft = 0.2748022651672363
Average tbt = 0.09020151719450951
P50 tbt = 0.09509240388870241
P99 tbt = 0.3038358302116394
All GPUs and memories are cold after 25.041831731796265
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.14754468447541538
P50 ttft = 0.11503934860229492
P99 ttft = 0.27917383193969725
Average tbt = 0.11722429517197286
P50 tbt = 0.09644706249237062
P99 tbt = 0.4494651489257814
All GPUs and memories are cold after 23.165935516357422
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1558666803750647
P50 ttft = 0.12389755249023438
P99 ttft = 0.3217507648468015
Average tbt = 0.11860464480986083
P50 tbt = 0.09592015743255616
P99 tbt = 0.4828920392990108
All GPUs and memories are cold after 24.041587591171265
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.11585126320521037
P50 ttft = 0.10648572444915771
P99 ttft = 0.19671750068664556
Average tbt = 0.03795913855234782
P50 tbt = 0.0231217622756958
P99 tbt = 0.12374391365051271
All GPUs and memories are cold after 2.0683906078338623
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1156880742027646
P50 ttft = 0.10626626014709473
P99 ttft = 0.20204644203186037
Average tbt = 0.040482909338814876
P50 tbt = 0.02317378520965576
P99 tbt = 0.15443969726562506
All GPUs and memories are cold after 5.051022529602051
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.12428937639508929
P50 ttft = 0.10622978210449219
P99 ttft = 0.25040792942047096
Average tbt = 0.05768069335392544
P50 tbt = 0.023592758178710937
P99 tbt = 0.1880128254890441
All GPUs and memories are cold after 14.10171389579773
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.12840533838039492
P50 ttft = 0.10681676864624023
P99 ttft = 0.25328302383422857
Average tbt = 0.06531644332699661
P50 tbt = 0.023568439483642577
P99 tbt = 0.28318285465240495
All GPUs and memories are cold after 16.027127504348755
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.13439635753631593
P50 ttft = 0.10609829425811768
P99 ttft = 0.2521806502342223
Average tbt = 0.07725739431381226
P50 tbt = 0.024678349494934082
P99 tbt = 0.28263224506378165
All GPUs and memories are cold after 19.031347513198853
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1411726474761963
P50 ttft = 0.1121673583984375
P99 ttft = 0.27487672567367555
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000001:00:00.0
Warning: persistence mode is disabled on device 00000001:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000002:00:00.0
Warning: persistence mode is disabled on device 00000002:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000003:00:00.0
Warning: persistence mode is disabled on device 00000003:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000008:00:00.0
Warning: persistence mode is disabled on device 00000008:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 00000009:00:00.0
Warning: persistence mode is disabled on device 00000009:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000A:00:00.0
Warning: persistence mode is disabled on device 0000000A:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000B:00:00.0
Warning: persistence mode is disabled on device 0000000B:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
GPU clocks set to "(gpuClkMin 1980, gpuClkMax 1980)" for GPU 0000000C:00:00.0
Warning: persistence mode is disabled on device 0000000C:00:00.0. See the Known Issues section of the nvidia-smi(1) man page for more information. Run with [--help | -h] switch to get more information on how to enable persistence mode.
All done.
All done.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...
To disable this warning, you can either:
- Avoid using `tokenizers` before the fork if possible
- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)
Average tbt = 0.09030852727591995
P50 tbt = 0.09562392234802247
P99 tbt = 0.306017765045166
All GPUs and memories are cold after 22.042361736297607
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.14711293129071798
P50 ttft = 0.11532425880432129
P99 ttft = 0.2798439598083496
Average tbt = 0.11684836720767089
P50 tbt = 0.0960063695907593
P99 tbt = 0.44891474246978774
All GPUs and memories are cold after 26.045918464660645
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1563370888491711
P50 ttft = 0.12442755699157715
P99 ttft = 0.32085416316986054
Average tbt = 0.11907833311931199
P50 tbt = 0.09595875740051271
P99 tbt = 0.4829485316276546
All GPUs and memories are cold after 26.053284168243408
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.11533067623774211
P50 ttft = 0.10614264011383057
P99 ttft = 0.1952427434921265
Average tbt = 0.03787365953127544
P50 tbt = 0.02320209741592407
P99 tbt = 0.12354113864898685
All GPUs and memories are cold after 2.011094808578491
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.11586594581604004
P50 ttft = 0.10657715797424316
P99 ttft = 0.2012669086456299
Average tbt = 0.04046597594306582
P50 tbt = 0.023264336585998534
P99 tbt = 0.1538731861114503
All GPUs and memories are cold after 8.072475910186768
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.12527217183794295
P50 ttft = 0.10683727264404297
P99 ttft = 0.2518232107162473
Average tbt = 0.05776239395141602
P50 tbt = 0.023696708679199218
P99 tbt = 0.18878099155426017
All GPUs and memories are cold after 16.06547474861145
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1295334246100449
P50 ttft = 0.10729002952575684
P99 ttft = 0.25274858474731454
Average tbt = 0.06547618028594228
P50 tbt = 0.023739075660705565
P99 tbt = 0.284424777030945
All GPUs and memories are cold after 16.030409812927246
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.13515560150146486
P50 ttft = 0.10664808750152588
P99 ttft = 0.2545101499557494
Average tbt = 0.07736738777160646
P50 tbt = 0.02474541664123535
P99 tbt = 0.283922440290451
All GPUs and memories are cold after 20.034026384353638
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.14004628732800484
P50 ttft = 0.11144816875457764
P99 ttft = 0.27636913537979124
Average tbt = 0.0898672714829445
P50 tbt = 0.09484088420867923
P99 tbt = 0.3047203702926636
All GPUs and memories are cold after 23.09816074371338
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.1460604145102305
P50 ttft = 0.1147456169128418
P99 ttft = 0.2748136138916016
Average tbt = 0.11650552880274108
P50 tbt = 0.09592332839965823
P99 tbt = 0.4481335563659669
All GPUs and memories are cold after 25.042633771896362
92886
288 - 96
288 - 256
288 - 1024
1053 - 5
1053 - 256
1053 - 600
8170 - 5
8170 - 256
8170 - 600
1053 - 5
Average ttft = 0.15620381860847932
P50 ttft = 0.12440085411071777
P99 ttft = 0.3229329061508177
Average tbt = 0.11899615511836778
P50 tbt = 0.09635846614837648
P99 tbt = 0.4844228057861324
All GPUs and memories are cold after 29.046589612960815
Exception ignored in: <module 'threading' from '/usr/lib/python3.8/threading.py'>
Traceback (most recent call last):
File "/usr/lib/python3.8/threading.py", line 1388, in _shutdown
lock.acquire()
KeyboardInterrupt: