Skip to content

Commit 5836136

Browse files
committed
optimiztions for serving
Signed-off-by: Tsai, Louie <[email protected]>
1 parent ee193b7 commit 5836136

File tree

1 file changed

+10
-0
lines changed

1 file changed

+10
-0
lines changed

.buildkite/nightly-benchmarks/tests/serving-tests-cpu.json

Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -6,6 +6,7 @@
66
"VLLM_RPC_TIMEOUT": 100000,
77
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
88
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
9+
"VLLM_CPU_NUM_OF_RESERVED_CPU": 0,
910
"VLLM_CPU_KVCACHE_SPACE": 40
1011
},
1112
"server_parameters": {
@@ -18,6 +19,7 @@
1819
"trust_remote_code": "",
1920
"disable_log_stats": "",
2021
"disable_log_requests": "",
22+
"enforce_eager": "",
2123
"load_format": "dummy"
2224
},
2325
"client_parameters": {
@@ -35,6 +37,7 @@
3537
"VLLM_RPC_TIMEOUT": 100000,
3638
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
3739
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
40+
"VLLM_CPU_NUM_OF_RESERVED_CPU": 0,
3841
"VLLM_CPU_KVCACHE_SPACE": 40
3942
},
4043
"server_parameters": {
@@ -47,6 +50,7 @@
4750
"trust_remote_code": "",
4851
"disable_log_stats": "",
4952
"disable_log_requests": "",
53+
"enforce_eager": "",
5054
"load_format": "dummy"
5155
},
5256
"client_parameters": {
@@ -64,6 +68,7 @@
6468
"VLLM_RPC_TIMEOUT": 100000,
6569
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
6670
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
71+
"VLLM_CPU_NUM_OF_RESERVED_CPU": 0,
6772
"VLLM_CPU_KVCACHE_SPACE": 40
6873
},
6974
"server_parameters": {
@@ -76,6 +81,7 @@
7681
"trust_remote_code": "",
7782
"disable_log_stats": "",
7883
"disable_log_requests": "",
84+
"enforce_eager": "",
7985
"load_format": "dummy"
8086
},
8187
"client_parameters": {
@@ -93,6 +99,7 @@
9399
"VLLM_RPC_TIMEOUT": 100000,
94100
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
95101
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
102+
"VLLM_CPU_NUM_OF_RESERVED_CPU": 0,
96103
"VLLM_CPU_KVCACHE_SPACE": 40
97104
},
98105
"server_parameters": {
@@ -106,6 +113,7 @@
106113
"enable_chunked_prefill": "",
107114
"disable_log_stats": "",
108115
"disable_log_requests": "",
116+
"enforce_eager": "",
109117
"load_format": "dummy"
110118
},
111119
"client_parameters": {
@@ -125,6 +133,7 @@
125133
"VLLM_RPC_TIMEOUT": 100000,
126134
"VLLM_ALLOW_LONG_MAX_MODEL_LEN": 1,
127135
"VLLM_ENGINE_ITERATION_TIMEOUT_S": 120,
136+
"VLLM_CPU_NUM_OF_RESERVED_CPU": 0,
128137
"VLLM_CPU_KVCACHE_SPACE": 40
129138
},
130139
"server_parameters": {
@@ -138,6 +147,7 @@
138147
"enable_chunked_prefill": "",
139148
"disable_log_stats": "",
140149
"disable_log_requests": "",
150+
"enforce_eager": "",
141151
"load_format": "dummy"
142152
},
143153
"client_parameters": {

0 commit comments

Comments
 (0)