forked from llm-d/llm-d-benchmark
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathbr_v0_1_example.yaml
More file actions
276 lines (276 loc) · 7.36 KB
/
Copy pathbr_v0_1_example.yaml
File metadata and controls
276 lines (276 loc) · 7.36 KB
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
version: '0.1' # Apply a version that updates with schema changes
scenario: # This section provides the specific environment and workload
description: This is a heterogeneous accelerator setup with two lora adapters
host:
type: # This will either be all "replica" or a mix of "prefill" and "decode"
- prefill
- decode
- decode
accelerator: # This is heterogeneous across prefill and decode, with 1 prefill and 2 decode (defined in scenario.host.type)
- model: H100 # Prefill
memory: 80
count: 1
parallelism:
dp: 1
tp: 1
pp: 1
ep: 1
- model: H100 # First decode
memory: 80
count: 8
parallelism:
dp: 1
tp: 8
pp: 1
ep: 8
- model: H100 # Second decode
memory: 80
count: 8
parallelism:
dp: 1
tp: 8
pp: 1
ep: 8
platform:
engine: # This list correlates 1:1 with the items listed in scenario.host.accelerator
- name: vllm # Prefill
version: 0.9.0.1
args:
"--dtype": fp16
"--tensor-parallel-size": 1
"--pipeline-parallel-size": 1
"--enable-expert-parallel": true
"--data-parallel-size": 1
"--data-parallel-size-local": 1
- name: vllm # First decode
version: 0.9.0.1
args:
"--dtype": fp16
"--tensor-parallel-size": 8
"--pipeline-parallel-size": 1
"--enable-expert-parallel": true
"--data-parallel-size": 3
"--data-parallel-size-local": 1
"--data-parallel-address": 10.12.33.212
"--data-parallel-rpc-port": 5555
"--data-parallel-start-rank": 1
- name: vllm # Second decode
version: 0.9.0.1
args:
"--dtype": fp16
"--tensor-parallel-size": 8
"--pipeline-parallel-size": 1
"--enable-expert-parallel": true
"--data-parallel-size": 3
"--data-parallel-size-local": 1
"--data-parallel-address": 10.12.33.212
"--data-parallel-rpc-port": 5555
"--data-parallel-start-rank": 2
model:
name: deepseek-ai/DeepSeek-R1-0528
quantization: fp16
adapters:
- lora: sql_adapter
- lora: golang_adapter
load:
name: inference-perf
type: long-input
args: # This section is currently unique to each harness. If this can be standardized, it may serve as a universal input to launching benchmark runs.
qps_values: 1.34
num_users_warmup: 20
num_users: 15
num_rounds: 20
system_prompt: 1000
chat_history: 20000
answer_len: 100
test_duration: 100
use_chat_completions: false
metrics: # These are the aggregate results from benchmarking
time:
duration: 16.531641244888306
start: 1749570583.5714512 # UTC seconds from epoch
stop: 1749570580.1030924
requests:
total: 32
failures: 0
incomplete: 1
input_length:
units: count
mean: 628.606060606061
stddev: 19.8353456345
min: 4
p10: 11
p50: 364
p90: 2427
max: 3836
output_length:
units: count
mean: 31.7878787878788
stddev: 19.8353456345
min: 30
p10: 31
p50: 32
p90: 32
max: 32
latency:
request_latency:
units: ms
mean: 3.31325431142327
stddev: 0.00198353456345
min: 1.62129471905064
p10: 1.67609986825846
p50: 2.11507539497688
p90: 5.94717199734878
max: 6.30658466403838
normalized_time_per_output_token:
units: ms/token
mean: 0.104340420636009
stddev: 0.00198353456345
min: 0.0506654599703325
p10: 0.0523781208830769
p50: 0.0670631669655753
p90: 0.189047570470012
max: 0.20343821496898
time_per_output_token:
units: ms/token
mean: 0.0836929455635872
stddev: 0.00198353456345
min: 0.0517028436646797
p10: 0.0530815053513894
p50: 0.0611870964678625
p90: 0.152292036800645
max: 0.17837208439984
time_to_first_token:
units: ms
mean: 0.800974442732916
stddev: 0.00198353456345
min: 0.0625283779809251
p10: 0.072068731742911
p50: 0.203539535985328
p90: 2.26959549135063
max: 4.46773961000145
inter_token_latency:
units: ms/token
mean: 0.0836929455635872
stddev: 0.00198353456345
min: 7.129972800612e-06
p10: 0.0534287681337446
p50: 0.0591336835059337
p90: 0.084046097996179
max: 0.614475268055685
throughput:
input_tokens_per_sec: 643.576644186323
output_tokens_per_sec: 32.544923821416
total_tokens_per_sec: 676.121568007739
requests_per_sec: 1.0238155253639
service: # These are metrics about the inference service
batch_size:
units: count
mean: 234.23049
stddev: 34.12342
min: 123
p10: 143
p50: 533
p90: 625
max: 753
queue_size:
units: count
mean: 234.12451
stddev: 34.56737
min: 123
p10: 143
p50: 533
p90: 625
max: 753
kv_cache_size:
units: count
mean: 2194993.253
stddev: 2342.3456
min: 1194345
p10: 1394456
p50: 2404751
p90: 2534437
max: 2554393
resources: # These are hardware level metrics
accelerator: # This list correlates 1:1 with the items listed in scenario.host.accelerator
- memory: # This corresponds to the prefill pod
consumption:
units: MB
mean: 2194993.2346
stddev: 2342.4568
min: 1194345
p10: 1394456
p50: 2404751
p90: 2534437
max: 2554393
utilization:
units: percent
mean: 80.235
stddev: 32.1
min: 40.3
p10: 44.4
p50: 71.3
p90: 97.1
max: 99.2
bandwidth:
units: MB/s
mean: 21993.2346
stddev: 22.4568
min: 19445.2347
p10: 13456.5367
p50: 24051.2456
p90: 24437.4582
max: 25543.3457
compute:
utilization:
units: percent
mean: 40.56
stddev: 12.15
min: 20.3
p10: 24.4
p50: 31.3
p90: 47.1
max: 49.2
power:
units: Watts
mean: 410.02
stddev: 170.1
min: 201.3
p10: 243.4
p50: 314.3
p90: 475.1
max: 497.2
- memory: # This corresponds to the first decode pod
consumption:
units: MB
mean: 2194993.2346
utilization:
units: percent
mean: 80.235
bandwidth:
units: MB/s
mean: 21993.2346
compute:
utilization:
units: percent
mean: 40.56
power:
units: Watts
mean: 410.02
- memory: # This corresponds to the second decode pod
consumption:
units: MB
mean: 2194993.2346
utilization:
units: percent
mean: 80.235
bandwidth:
units: MB/s
mean: 21993.2346
compute:
utilization:
units: percent
mean: 40.56
power:
units: Watts
mean: 410.02