Skip to content

Latest commit

 

History

History
160 lines (121 loc) · 5.87 KB

ensemble_quick_start.md

File metadata and controls

160 lines (121 loc) · 5.87 KB
Raw

Ensemble Model Quick Start

The steps below will guide you through using Model Analyzer in Docker mode to profile and analyze a simple ensemble model: ensemble_add_sub.

Step 1: Download the ensemble model ensemble_add_sub and composing models add, sub

1. Create a new directory and enter it

mkdir <new_dir> && cd <new_dir>

2. Start a git repository

git init && git remote add -f origin https://github.com/triton-inference-server/model_analyzer.git

3. Enable sparse checkout, and download the examples directory, which contains the ensemble_add_sub, add and sub

git config core.sparseCheckout true && \
echo 'examples' >> .git/info/sparse-checkout && \
git pull origin main

3. Add a version directory to ensemble_add_sub

mkdir examples/quick-start/ensemble_add_sub/1

Step 2: Pull and Run the SDK Container

1. Pull the SDK container:

docker pull nvcr.io/nvidia/tritonserver:24.11-py3-sdk

2. Run the SDK container

docker run -it --gpus 1 \
      --shm-size 1G \
      -v /var/run/docker.sock:/var/run/docker.sock \
      -v $(pwd)/examples/quick-start:$(pwd)/examples/quick-start \
      --net=host nvcr.io/nvidia/tritonserver:24.11-py3-sdk

Important: The example above uses a single GPU. If you are running on multiple GPUs, you may need to increase the shared memory size accordingly

Step 3: Profile the ensemble_add_sub model

The examples/quick-start directory is an example Triton Model Repository that contains the ensemble model ensemble_add_sub, which calculates the sum and difference of two inputs using add and sub models.

Run the Model Analyzer profile subcommand inside the container with:

model-analyzer profile \
    --model-repository <path-to-examples-quick-start> \
    --profile-models ensemble_add_sub \
    --triton-launch-mode=docker --triton-docker-shm-size=1G \
    --output-model-repository-path <path-to-output-model-repo>/<output_dir> \
    --export-path profile_results

Important: You must specify an <output_dir> subdirectory. You cannot have --output-model-repository-path point directly to <path-to-output-model-repo>

Important: If you already ran this earlier in the container, you can use the --override-output-model-repository option to overwrite the earlier results.

Important: All models must be in the same repository

The Model analyzer uses Quick Search algorithm for profiling the Ensemble model. After the quick search is completed, Model Analyzer will then sweep concurrencies for the top three configurations and then create a summary report and CSV outputs.

Here is an example result summary, run on a Tesla V100 GPU:

Result Summary Top Result Summary Table

You will note that the top model configuration has a higher throughput than the other configurations.

The measured data and summary report will be placed inside the ./profile_results directory. The directory will be structured as follows.

$HOME
|-- model_analyzer
    |-- profile_results
        |-- plots
        |   |-- detailed
        |   |   |-- ensemble_add_sub_config_5
        |   |   |   `-- latency_breakdown.png
        |   |   |-- ensemble_add_sub_config_6
        |   |   |   `-- latency_breakdown.png
        |   |   `-- ensemble_add_sub_config_7
        |   |       `-- latency_breakdown.png
        |   `-- simple
        |       |-- ensemble_add_sub
        |       |   |-- gpu_mem_v_latency.png
        |       |   `-- throughput_v_latency.png
        |       |-- ensemble_add_sub_config_5
        |       |   |-- cpu_mem_v_latency.png
        |       |   |-- gpu_mem_v_latency.png
        |       |   |-- gpu_power_v_latency.png
        |       |   `-- gpu_util_v_latency.png
        |       |-- ensemble_add_sub_config_6
        |       |   |-- cpu_mem_v_latency.png
        |       |   |-- gpu_mem_v_latency.png
        |       |   |-- gpu_power_v_latency.png
        |       |   `-- gpu_util_v_latency.png
        |       `-- ensemble_add_sub_config_7
        |           |-- cpu_mem_v_latency.png
        |           |-- gpu_mem_v_latency.png
        |           |-- gpu_power_v_latency.png
        |           `-- gpu_util_v_latency.png
        |-- reports
        |   |-- detailed
        |   |   |-- ensemble_add_sub_config_5
        |   |   |   `-- detailed_report.pdf
        |   |   |-- ensemble_add_sub_config_6
        |   |   |   `-- detailed_report.pdf
        |   |   `-- ensemble_add_sub_config_7
        |   |       `-- detailed_report.pdf
        |   `-- summaries
        |       `-- ensemble_add_sub
        |           `-- result_summary.pdf
        `-- results
            |-- metrics-model-gpu.csv
            |-- metrics-model-inference.csv
            `-- metrics-server-only.csv

Note: Above configurations, ensemble_add_sub_config_5, ensemble_add_sub_config_6, and ensemble_add_sub_config_7 are generated as the top configurations when running profiling on a single Tesla V100 GPU. However, running on multiple GPUs or different model GPUs may result in different top configurations.