Skip to content

Commit 660a8db

Browse files
committed
[Fix]Add chunk size in test cases
1 parent 0abf14b commit 660a8db

2 files changed

Lines changed: 5 additions & 0 deletions

File tree

examples/ucm_config_example.yaml

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -24,6 +24,8 @@ enable_event_sync: true
2424
# Enable UCM metrics so they can be monitored online via Grafana and Prometheus.
2525
# metrics_config_path: "/workspace/unified-cache-management/examples/metrics/metrics_configs.yaml"
2626

27+
chunk_size: 256
28+
2729
# Sparse attention configuration
2830
# ucm_sparse_config:
2931
# GSAOnDevice: {}

test/suites/E2E/test_online_inference.py

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -96,6 +96,7 @@ def test_online_accuracy_hbm_ssd_mixed(
9696
if ucm_connector_name == "UcmNfsStore"
9797
else {}
9898
),
99+
"chunk_size": 256,
99100
}
100101

101102
# Build vllm_server_startup_args
@@ -183,6 +184,7 @@ def test_online_accuracy_hbm_ssd_mixed_pp(
183184
if ucm_connector_name == "UcmNfsStore"
184185
else {}
185186
),
187+
"chunk_size": 256,
186188
}
187189

188190
# Build vllm_server_startup_args with pipeline parallel size
@@ -256,6 +258,7 @@ def test_online_accuracy_hbm_ssd_mixed_tp(
256258
if ucm_connector_name == "UcmNfsStore"
257259
else {}
258260
),
261+
"chunk_size": 256,
259262
}
260263

261264
# Build vllm_server_startup_args with tensor parallel size

0 commit comments

Comments
 (0)