Run (Debug) - spice_cloud #121
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Run (Debug - Spice Cloud) | |
| run-name: Run (Debug) - ${{ github.event.inputs.system_under_test || 'spice_cloud' }} | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| system_under_test: | |
| description: "System under test" | |
| required: true | |
| default: spice_cloud | |
| type: choice | |
| options: | |
| - spice_cloud | |
| environment: | |
| description: "Target Spice Cloud Environment" | |
| required: true | |
| default: "dev" | |
| type: choice | |
| options: | |
| - dev | |
| - prod | |
| scenario: | |
| description: "Scenario/query set to run (e.g. tpch)" | |
| required: true | |
| default: "tpch" | |
| type: string | |
| etl_type: | |
| description: "ETL type" | |
| required: true | |
| default: "events" | |
| type: choice | |
| options: | |
| - events | |
| - changes | |
| scale_factor: | |
| description: "Scale Factor" | |
| required: true | |
| default: "1" | |
| type: choice | |
| options: | |
| - "0.1" | |
| - "1" | |
| - "10" | |
| spidapter_version: | |
| description: "Spidapter image version tag (e.g. latest, v1.0.0)" | |
| required: false | |
| default: "latest" | |
| type: string | |
| num_query_clients: | |
| description: "Number of concurrent query clients" | |
| required: false | |
| default: "2" | |
| type: string | |
| enable_module_debug_logging: | |
| description: "Enable debug logs" | |
| required: false | |
| default: false | |
| type: boolean | |
| disable_teardown: | |
| description: "Skip the teardown RPC call to the system adapter" | |
| required: false | |
| default: false | |
| type: boolean | |
| enable_pvc: | |
| description: "Enable PVC block storage for scheduler and executor pods (sizes are derived from scale_factor)" | |
| required: false | |
| default: false | |
| type: boolean | |
| executor_replicas: | |
| description: "Number of executor replicas" | |
| required: false | |
| default: "4" | |
| type: string | |
| app_memory_limit: | |
| description: "Memory limit for the scheduler (app) pod (e.g. 16Gi, 20Gi)" | |
| required: false | |
| default: "16Gi" | |
| type: string | |
| executor_memory_limit: | |
| description: "Memory limit for the executor pod (e.g. 16Gi, 62Gi)" | |
| required: false | |
| default: "16Gi" | |
| type: string | |
| custom_image: | |
| description: "Custom runtime container image (e.g. ghcr.io/spiceai/spiceai-dev:spicebench-sf10). Overrides the default channel image. Requires the internal update channel on the target SCP environment." | |
| required: false | |
| default: "" | |
| type: string | |
| jobs: | |
| run-spicebench: | |
| name: Run spicebench | |
| runs-on: spiceai-dev-runners | |
| timeout-minutes: 600 | |
| steps: | |
| - uses: actions/checkout@v6 | |
| - uses: ./.github/actions/setup-cc | |
| - uses: ./.github/actions/management-login | |
| with: | |
| token-url: ${{ github.event.inputs.environment == 'prod' && 'https://spice.ai/api/oauth/token' || 'https://dev.spice.ai/api/oauth/token' }} | |
| client-id: ${{ github.event.inputs.environment == 'prod' && secrets.SPICE_MANAGEMENT_CLIENT_ID_PROD || secrets.SPICE_MANAGEMENT_CLIENT_ID }} | |
| client-secret: ${{ github.event.inputs.environment == 'prod' && secrets.SPICE_MANAGEMENT_CLIENT_SECRET_PROD || secrets.SPICE_MANAGEMENT_CLIENT_SECRET }} | |
| - name: Log in to GHCR | |
| uses: docker/login-action@v3 | |
| with: | |
| registry: ghcr.io | |
| username: ${{ github.actor }} | |
| password: ${{ secrets.GITHUB_TOKEN }} | |
| - name: pull spidapter image | |
| run: docker pull ghcr.io/spiceai/spidapter:${{ github.event.inputs.spidapter_version || 'latest' }} | |
| - uses: ./.github/actions/build-spicebench | |
| - name: Validate adapter configuration | |
| env: | |
| SCENARIO: ${{ github.event.inputs.scenario || 'tpch' }} | |
| SPICEAI_API_KEY: ${{ env.SPICEAI_API_KEY }} | |
| SPICE_CLOUD_API_URL: ${{ github.event.inputs.environment == 'prod' && 'https://api.spice.ai' || 'https://dev-api.spice.ai' }} | |
| run: | | |
| set -euo pipefail | |
| if [ -z "${SCENARIO}" ]; then | |
| echo "SCENARIO must not be empty" | |
| exit 1 | |
| fi | |
| if [ -z "${SPICEAI_API_KEY:-}" ]; then | |
| echo "SPICEAI_API_KEY must be set for spice_cloud" | |
| exit 1 | |
| fi | |
| if ! command -v docker >/dev/null 2>&1; then | |
| echo "docker is required for spice_cloud mode" | |
| exit 1 | |
| fi | |
| docker image inspect ghcr.io/spiceai/spidapter:${{ github.event.inputs.spidapter_version || 'latest' }} >/dev/null 2>&1 || { | |
| echo "spidapter docker image not found locally; pull step may have failed" | |
| exit 1 | |
| } | |
| - name: Install ADBC FlightSQL driver | |
| uses: columnar-tech/setup-dbc@v1 | |
| with: | |
| drivers: flightsql | |
| - name: Run spicebench | |
| env: | |
| SPICEAI_API_KEY: ${{ env.SPICEAI_API_KEY }} | |
| SPICE_CLOUD_API_URL: ${{ github.event.inputs.environment == 'prod' && 'https://api.spice.ai' || 'https://dev-api.spice.ai' }} | |
| SCENARIO: ${{ github.event.inputs.scenario || 'tpch' }} | |
| SYSTEM_UNDER_TEST: ${{ github.event.inputs.system_under_test || 'spice_cloud' }} | |
| SYSTEM_ADAPTER: ${{ github.event.inputs.system_under_test || 'spice_cloud' }} | |
| NUM_QUERY_CLIENTS: ${{ github.event.inputs.num_query_clients || '2' }} | |
| ETL_BUCKET: "spicebench" | |
| ETL_PREFIX: ${{ github.event.inputs.etl_type == 'changes' && 'data-gen-mutable' || 'data-gen' }} | |
| SCALE_FACTOR: ${{ github.event.inputs.scale_factor || '1' }} | |
| ETL_REGION: "us-east-1" | |
| ETL_SINK: "adbc" | |
| SCHEDULER_STATE_LOCATION: "s3://spiceai-testing-cluster-state/spicebench-scheduler-state-${{ github.run_id }}/" | |
| VALIDATE_CHECKPOINT_RESULTS: "true" | |
| ENABLE_MODULE_DEBUG_LOGGING: ${{ github.event.inputs.enable_module_debug_logging || 'false' }} | |
| SCRAPE_SUT_METRICS: "true" | |
| SPICEAI_BENCHMARK_METRICS_KEY: ${{ secrets.SPICEAI_BENCHMARK_METRICS_KEY }} | |
| MINIO_ENDPOINT: ${{ secrets.MINIO_ENDPOINT }} | |
| AWS_ACCESS_KEY_ID: ${{ secrets.MINIO_ACCESS_KEY_ID }} | |
| AWS_SECRET_ACCESS_KEY: ${{ secrets.MINIO_SECRET_ACCESS_KEY }} | |
| S3_AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }} | |
| S3_AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }} | |
| SPIDAPTER_ICEBERG_REGION: us-west-1 | |
| SPIDAPTER_ICEBERG_CATALOG_FROM: iceberg:https://glue.us-west-1.amazonaws.com/iceberg/v1/catalogs/211125479522/namespaces | |
| DISABLE_TEARDOWN: ${{ github.event.inputs.disable_teardown || 'false' }} | |
| ENABLE_PVC: ${{ github.event.inputs.enable_pvc || 'false' }} | |
| EXECUTOR_REPLICAS: ${{ github.event.inputs.executor_replicas || '4' }} | |
| SPIDAPTER_APP_MEMORY_LIMIT: ${{ github.event.inputs.app_memory_limit || '16Gi' }} | |
| SPIDAPTER_EXECUTOR_MEMORY_LIMIT: ${{ github.event.inputs.executor_memory_limit || '16Gi' }} | |
| SPIDAPTER_ORGANIZATION_TAG: "spicehq" | |
| CUSTOM_IMAGE: ${{ github.event.inputs.custom_image || '' }} | |
| run: | | |
| set -euo pipefail | |
| if [ "${ENABLE_MODULE_DEBUG_LOGGING}" = "true" ]; then | |
| export RUST_LOG='info,etl=debug,spicebench=debug,data_generation=debug' | |
| else | |
| export RUST_LOG='info' | |
| fi | |
| TABLE_FORMAT="parquet" | |
| EXECUTOR_INSTANCE_TYPE="github-hosted-ubuntu-latest" | |
| ETL_ENDPOINT="${MINIO_ENDPOINT}" | |
| ETL_ARGS="--etl-bucket ${ETL_BUCKET} --scale-factor ${SCALE_FACTOR}" | |
| if [ -n "${ETL_PREFIX}" ]; then | |
| ETL_ARGS="${ETL_ARGS} --etl-prefix ${ETL_PREFIX}" | |
| fi | |
| if [ -n "${ETL_REGION}" ]; then | |
| ETL_ARGS="${ETL_ARGS} --etl-region ${ETL_REGION}" | |
| fi | |
| if [ -n "${ETL_ENDPOINT:-}" ]; then | |
| ETL_ARGS="${ETL_ARGS} --etl-endpoint ${ETL_ENDPOINT}" | |
| fi | |
| ETL_SINK_ARGS="--etl-sink ${ETL_SINK} --table-format ${TABLE_FORMAT}" | |
| if [ "${ETL_SINK}" = "adbc" ]; then | |
| : | |
| fi | |
| VALIDATION_ARGS="" | |
| if [ "${VALIDATE_CHECKPOINT_RESULTS}" = "true" ]; then | |
| VALIDATION_ARGS="--validate-results" | |
| fi | |
| SCHEDULER_STATE_ADAPTER_ENV="--system-adapter-env SCHEDULER_STATE_LOCATION=${SCHEDULER_STATE_LOCATION}" | |
| SUT_METRICS_ARGS="" | |
| if [ "${SCRAPE_SUT_METRICS}" = "true" ]; then | |
| SUT_METRICS_ARGS="--scrape-sut-metrics" | |
| fi | |
| export SPICEBENCH_ADBC_UPDATE_STRATEGY=bulk_ingest_upsert | |
| export SPICEBENCH_ADBC_FLUSH_STREAM_BEFORE_UPSERT=true | |
| export SPICEBENCH_ADBC_DELETE_BATCH_SIZE=50000 | |
| export SPICEBENCH_ADBC_SINK_POOL_SIZE=1 | |
| ADAPTER_CMD="docker" | |
| ADAPTER_DOCKER_OPTS="run -i -e SPIDAPTER_EXECUTOR_REPLICAS=${EXECUTOR_REPLICAS} -e SPICEAI_API_KEY -e SPICE_CLOUD_API_URL -e AWS_ACCESS_KEY_ID=${S3_AWS_ACCESS_KEY_ID} -e AWS_SECRET_ACCESS_KEY=${S3_AWS_SECRET_ACCESS_KEY} -e SPIDAPTER_ICEBERG_REGION -e SPIDAPTER_ICEBERG_CATALOG_FROM -e SCHEDULER_STATE_LOCATION" | |
| ADAPTER_DOCKER_OPTS="${ADAPTER_DOCKER_OPTS} -e SPIDAPTER_APP_MEMORY_LIMIT=${SPIDAPTER_APP_MEMORY_LIMIT}" | |
| ADAPTER_DOCKER_OPTS="${ADAPTER_DOCKER_OPTS} -e SPIDAPTER_EXECUTOR_MEMORY_LIMIT=${SPIDAPTER_EXECUTOR_MEMORY_LIMIT}" | |
| ADAPTER_DOCKER_OPTS="${ADAPTER_DOCKER_OPTS} -e SPIDAPTER_ORGANIZATION_TAG" | |
| # Parse custom image into registry, name, and tag for spidapter. | |
| # Expected format: registry/image:tag (e.g. ghcr.io/spiceai/spiceai-dev:spicebench-sf10) | |
| # Parses as: registry=ghcr.io, image=spiceai/spiceai-dev, tag=spicebench-sf10 | |
| if [ -n "${CUSTOM_IMAGE}" ]; then | |
| IMAGE_WITH_TAG="${CUSTOM_IMAGE}" | |
| if [[ "${IMAGE_WITH_TAG}" == *":"* ]]; then | |
| SPIDAPTER_IMAGE_TAG="${IMAGE_WITH_TAG##*:}" | |
| IMAGE_WITHOUT_TAG="${IMAGE_WITH_TAG%:*}" | |
| else | |
| SPIDAPTER_IMAGE_TAG="" | |
| IMAGE_WITHOUT_TAG="${IMAGE_WITH_TAG}" | |
| fi | |
| # Registry is the first path component (hostname), image is the rest | |
| SPIDAPTER_IMAGE_REGISTRY="${IMAGE_WITHOUT_TAG%%/*}" | |
| SPIDAPTER_IMAGE_NAME="${IMAGE_WITHOUT_TAG#*/}" | |
| echo "Custom image: registry=${SPIDAPTER_IMAGE_REGISTRY}, name=${SPIDAPTER_IMAGE_NAME}, tag=${SPIDAPTER_IMAGE_TAG}" | |
| ADAPTER_DOCKER_OPTS="${ADAPTER_DOCKER_OPTS} -e SPIDAPTER_IMAGE_REGISTRY=${SPIDAPTER_IMAGE_REGISTRY} -e SPIDAPTER_IMAGE_NAME=${SPIDAPTER_IMAGE_NAME} -e SPIDAPTER_QUERY_MEMORY_LIMIT=500Gi" | |
| if [ -n "${SPIDAPTER_IMAGE_TAG}" ]; then | |
| ADAPTER_DOCKER_OPTS="${ADAPTER_DOCKER_OPTS} -e SPIDAPTER_IMAGE_TAG=${SPIDAPTER_IMAGE_TAG}" | |
| fi | |
| fi | |
| if [ "${ENABLE_PVC}" = "true" ]; then | |
| echo "PVC enabled: app=3GB, executor=2GB" | |
| ADAPTER_DOCKER_OPTS="${ADAPTER_DOCKER_OPTS} -e SPIDAPTER_APP_STORAGE_SIZE_GB=3 -e SPIDAPTER_EXECUTOR_STORAGE_SIZE_GB=2 -e SPIDAPTER_CAYENNE_DATA_DIR=/data/data -e SPIDAPTER_CAYENNE_METADATA_DIR=/data/metadata" | |
| fi | |
| # Use internal channel when a custom image is specified, otherwise nightly. | |
| SPIDAPTER_CHANNEL="nightly" | |
| if [ -n "${CUSTOM_IMAGE}" ]; then | |
| SPIDAPTER_CHANNEL="internal" | |
| fi | |
| ADAPTER_ARGS="${ADAPTER_DOCKER_OPTS} ghcr.io/spiceai/spidapter:${{ github.event.inputs.spidapter_version || 'latest' }} stdio --verbose --channel ${SPIDAPTER_CHANNEL}" | |
| ADAPTER_ENVS="" | |
| NO_TEARDOWN_ARG="" | |
| if [ "${DISABLE_TEARDOWN}" = "true" ]; then | |
| NO_TEARDOWN_ARG="--no-teardown" | |
| fi | |
| ~/.spice/bin/spicebench run \ | |
| --concurrency "${NUM_QUERY_CLIENTS}" \ | |
| --scenario "${SCENARIO}" \ | |
| --executor-instance-type "${EXECUTOR_INSTANCE_TYPE}" \ | |
| ${ETL_ARGS} \ | |
| ${ETL_SINK_ARGS} \ | |
| ${VALIDATION_ARGS} \ | |
| ${SUT_METRICS_ARGS} \ | |
| --system-adapter-stdio-cmd "${ADAPTER_CMD}" \ | |
| --system-adapter-stdio-args "${ADAPTER_ARGS}" \ | |
| ${ADAPTER_ENVS} \ | |
| ${SCHEDULER_STATE_ADAPTER_ENV} \ | |
| ${NO_TEARDOWN_ARG} \ |