Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Support inf request rate and number prompts in LPG #847

Merged
merged 3 commits into from
Oct 15, 2024
Merged

Support inf request rate and number prompts in LPG #847

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
f5f61bd
minor updates to lpg and checkpoint conversion
vivianrwu Oct 7, 2024
728e283
Support inf request rate and number prompts
vivianrwu Oct 9, 2024
da99beb
Merge branch 'GoogleCloudPlatform:main' into updates
vivianrwu Oct 9, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 8 additions & 1 deletion benchmarks/benchmark/tools/profile-generator/container/latency_throughput_curve.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,14 @@ for request_rate in $(echo $REQUEST_RATES | tr ',' ' '); do
# TODO: Check if profile already exists, if so then skip
timestamp=$(date +"%Y-%m-%d_%H-%M-%S")
output_file="latency-profile-${timestamp}.txt"
PYTHON_OPTS="$PYTHON_OPTS --save-json-results --host=$IP --port=$PORT --model=$TOKENIZER --dataset=$PROMPT_DATASET_FILE --tokenizer=$TOKENIZER --request-rate=$request_rate --backend=$BACKEND --num-prompts=$((request_rate * $BENCHMARK_TIME_SECONDS)) --max-input-length=$INPUT_LENGTH --max-output-length=$OUTPUT_LENGTH"
if [ ${request_rate} == 0 ]; then
request_rate="inf"
NUM_PROMPTS=$MAX_NUM_PROMPTS
else
NUM_PROMPTS=$((${request_rate} * $BENCHMARK_TIME_SECONDS))
fi

PYTHON_OPTS="$PYTHON_OPTS --save-json-results --host=$IP --port=$PORT --model=$TOKENIZER --dataset=$PROMPT_DATASET_FILE --tokenizer=$TOKENIZER --request-rate=$request_rate --backend=$BACKEND --num-prompts=$NUM_PROMPTS --max-input-length=$INPUT_LENGTH --max-output-length=$OUTPUT_LENGTH"
if [[ "$SCRAPE_SERVER_METRICS" = "true" ]]; then
PYTHON_OPTS="$PYTHON_OPTS --scrape-server-metrics"
fi
Expand Down
2 changes: 2 additions & 0 deletions ...e-generator/modules/latency-profile/manifest-templates/latency-profile-generator.yaml.tpl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,6 +42,8 @@ spec:
value: ${output_bucket}
- name: SCRAPE_SERVER_METRICS
value: ${scrape_server_metrics}
- name: MAX_NUM_PROMPTS
value: ${max_num_prompts}
%{ for hugging_face_token_secret in hugging_face_token_secret_list ~}
- name: HF_TOKEN
valueFrom:
Expand Down