diff --git a/docker-compose.yaml b/docker-compose.yaml index fede5634..c90b36a4 100644 --- a/docker-compose.yaml +++ b/docker-compose.yaml @@ -19,13 +19,13 @@ services: - NIM_HTTP_API_PORT=8000 - NIM_TRITON_LOG_VERBOSE=1 - NGC_API_KEY=${NIM_NGC_API_KEY:-${NGC_API_KEY:-ngcapikey}} - - CUDA_VISIBLE_DEVICES=1 + - CUDA_VISIBLE_DEVICES=0 deploy: resources: reservations: devices: - driver: nvidia - count: all + device_ids: ["1"] capabilities: [gpu] runtime: nvidia @@ -46,7 +46,7 @@ services: reservations: devices: - driver: nvidia - count: all + device_ids: ["0"] capabilities: [gpu] runtime: nvidia @@ -62,13 +62,13 @@ services: - NIM_HTTP_API_PORT=8000 - NIM_TRITON_LOG_VERBOSE=1 - NGC_API_KEY=${NIM_NGC_API_KEY:-${NGC_API_KEY:-ngcapikey}} - - CUDA_VISIBLE_DEVICES=1 + - CUDA_VISIBLE_DEVICES=0 deploy: resources: reservations: devices: - driver: nvidia - count: all + device_ids: ["1"] capabilities: [gpu] runtime: nvidia @@ -84,13 +84,13 @@ services: - NIM_HTTP_API_PORT=8000 - NIM_TRITON_LOG_VERBOSE=1 - NGC_API_KEY=${NIM_NGC_API_KEY:-${NGC_API_KEY:-ngcapikey}} - - CUDA_VISIBLE_DEVICES=1 + - CUDA_VISIBLE_DEVICES=0 deploy: resources: reservations: devices: - driver: nvidia - count: all + device_ids: ["1"] capabilities: [gpu] runtime: nvidia @@ -106,13 +106,13 @@ services: - NIM_HTTP_API_PORT=8000 - NIM_TRITON_LOG_VERBOSE=1 - NGC_API_KEY=${NIM_NGC_API_KEY:-${NGC_API_KEY:-ngcapikey}} - - CUDA_VISIBLE_DEVICES=1 + - CUDA_VISIBLE_DEVICES=0 deploy: resources: reservations: devices: - driver: nvidia - count: all + device_ids: ["1"] capabilities: [gpu] runtime: nvidia @@ -167,9 +167,10 @@ services: reservations: devices: - driver: nvidia - count: all + device_ids: ["1"] capabilities: [gpu] - + depends_on: + - milvus otel-collector: image: otel/opentelemetry-collector-contrib:0.91.0 @@ -258,32 +259,39 @@ services: # timeout: 20s # retries: 3 -# milvus: -# # Turn on to leverage the `vdb_upload` task -# restart: always -# container_name: milvus-standalone -# image: milvusdb/milvus:v2.3.5 -# command: ["milvus", "run", "standalone"] -# hostname: milvus -# security_opt: -# - seccomp:unconfined -# environment: -# ETCD_ENDPOINTS: etcd:2379 -# MINIO_ADDRESS: minio:9000 -# volumes: -# - ./.volumes/milvus:/var/lib/milvus -# healthcheck: -# test: ["CMD", "curl", "-f", "http://localhost:9091/healthz"] -# interval: 30s -# start_period: 90s -# timeout: 20s -# retries: 3 -# ports: -# - "19530:19530" -# - "9091:9091" -# depends_on: -# - "etcd" -# - "minio" + # milvus: + # # Turn on to leverage the `vdb_upload` task + # restart: always + # container_name: milvus-standalone + # image: milvusdb/milvus:v2.4.9-gpu + # command: ["milvus", "run", "standalone"] + # hostname: milvus + # security_opt: + # - seccomp:unconfined + # environment: + # ETCD_ENDPOINTS: etcd:2379 + # MINIO_ADDRESS: minio:9000 + # volumes: + # - ./.volumes/milvus:/var/lib/milvus + # healthcheck: + # test: ["CMD", "curl", "-f", "http://localhost:9091/healthz"] + # interval: 30s + # start_period: 90s + # timeout: 20s + # retries: 3 + # ports: + # - "19530:19530" + # - "9091:9091" + # deploy: + # resources: + # reservations: + # devices: + # - driver: nvidia + # device_ids: ["1"] + # capabilities: [gpu] + # depends_on: + # - "etcd" + # - "minio" # attu: # # Turn on to leverage the `vdb_upload` task diff --git a/src/nv_ingest/schemas/vdb_task_sink_schema.py b/src/nv_ingest/schemas/vdb_task_sink_schema.py index 74df980e..231c62bb 100644 --- a/src/nv_ingest/schemas/vdb_task_sink_schema.py +++ b/src/nv_ingest/schemas/vdb_task_sink_schema.py @@ -37,10 +37,11 @@ def build_default_milvus_config(embedding_size: int = 1024) -> typing.Dict[str, "index_conf": { "field_name": "vector", "metric_type": "L2", - "index_type": "HNSW", + "index_type": "GPU_CAGRA", "params": { - "M": 8, - "efConstruction": 64, + 'intermediate_graph_degree':128, + 'graph_degree': 64, + "build_algo": "NN_DESCENT", }, }, "schema_conf": {