GenAIExamples/GraphRAG/docker_compose/intel/hpu/gaudi/compose.yaml

# Copyright (C) 2024 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

services:
  neo4j-apoc:
    image: neo4j:latest
    container_name: neo4j-apoc
    volumes:
        - /$HOME/neo4j/logs:/logs
        - /$HOME/neo4j/config:/config
        - /$HOME/neo4j/data:/data
        - /$HOME/neo4j/plugins:/plugins
    ipc: host
    environment:
      - NEO4J_AUTH=${NEO4J_USERNAME}/${NEO4J_PASSWORD}
      - NEO4J_PLUGINS=["apoc"]
      - NEO4J_apoc_export_file_enabled=true
      - NEO4J_apoc_import_file_enabled=true
      - NEO4J_apoc_import_file_use__neo4j__config=true
      - NEO4J_dbms_security_procedures_unrestricted=apoc.\*
    ports:
      - "7474:7474"
      - "7687:7687"
    restart: always
  tei-embedding-service:
    image: ghcr.io/huggingface/text-embeddings-inference:cpu-1.5
    container_name: tei-embedding-server
    ports:
      - "6006:80"
    volumes:
      - "./data:/data"
    shm_size: 1g
    environment:
      no_proxy: ${no_proxy}
      NO_PROXY: ${no_proxy}
      http_proxy: ${http_proxy}
      https_proxy: ${https_proxy}
      HUGGING_FACE_HUB_TOKEN: ${HUGGINGFACEHUB_API_TOKEN}
    ipc: host
    command: --model-id ${EMBEDDING_MODEL_ID} --auto-truncate
  tgi-gaudi-service:
    image: ghcr.io/huggingface/tgi-gaudi:2.0.6
    container_name: tgi-gaudi-server
    ports:
      - "6005:80"
    volumes:
      - "./data:/data"
    environment:
      no_proxy: ${no_proxy}
      NO_PROXY: ${no_proxy}
      http_proxy: ${http_proxy}
      https_proxy: ${https_proxy}
      HUGGING_FACE_HUB_TOKEN: ${HUGGINGFACEHUB_API_TOKEN}
      HF_TOKEN: ${HF_TOKEN}
      HF_HUB_DISABLE_PROGRESS_BARS: 1
      HF_HUB_ENABLE_HF_TRANSFER: 0
      HABANA_VISIBLE_DEVICES: all
      OMPI_MCA_btl_vader_single_copy_mechanism: none
      ENABLE_HPU_GRAPH: true
      LIMIT_HPU_GRAPH: true
      USE_FLASH_ATTENTION: true
      FLASH_ATTENTION_RECOMPUTE: true
      TEXT_GENERATION_SERVER_IGNORE_EOS_TOKEN: false
    runtime: habana
    cap_add:
      - SYS_NICE
    ipc: host
    command: --model-id ${LLM_MODEL_ID} --max-input-length 6000 --max-total-tokens 8192
  dataprep-neo4j-llamaindex:
    image: ${REGISTRY:-opea}/dataprep:${TAG:-latest}
    container_name: dataprep-neo4j-server
    depends_on:
      - neo4j-apoc
      - tgi-gaudi-service
      - tei-embedding-service
    ports:
      - "6004:5000"
    ipc: host
    environment:
      no_proxy: ${no_proxy}
      http_proxy: ${http_proxy}
      https_proxy: ${https_proxy}
      host_ip: ${host_ip}
      HUGGING_FACE_HUB_TOKEN: ${HUGGINGFACEHUB_API_TOKEN}
      HF_TOKEN: ${HF_TOKEN}
      NEO4J_URL: ${NEO4J_URL}
      NEO4J_USERNAME: ${NEO4J_USERNAME}
      NEO4J_PASSWORD: ${NEO4J_PASSWORD}
      DATAPREP_COMPONENT_NAME: "OPEA_DATAPREP_NEO4J_LLAMAINDEX"
      TGI_LLM_ENDPOINT: ${TGI_LLM_ENDPOINT}
      TEI_EMBEDDING_ENDPOINT: ${TEI_EMBEDDING_ENDPOINT}
      OPENAI_API_KEY: ${OPENAI_API_KEY}
      OPENAI_EMBEDDING_MODEL: ${OPENAI_EMBEDDING_MODEL}
      OPENAI_LLM_MODEL: ${OPENAI_LLM_MODEL}
      EMBEDDING_MODEL_ID: ${EMBEDDING_MODEL_ID}
      LLM_MODEL_ID: ${LLM_MODEL_ID}
      MAX_OUTPUT_TOKENS: ${MAX_OUTPUT_TOKENS}
      LOGFLAG: ${LOGFLAG}
    restart: unless-stopped
  retriever-neo4j-llamaindex:
    image: ${REGISTRY:-opea}/retriever:${TAG:-latest}
    container_name: retriever-neo4j-server
    depends_on:
      - neo4j-apoc
      - tgi-gaudi-service
      - tei-embedding-service
    ports:
      - "7000:7000"
    ipc: host
    environment:
      no_proxy: ${no_proxy}
      http_proxy: ${http_proxy}
      https_proxy: ${https_proxy}
      host_ip: ${host_ip}
      HUGGING_FACE_HUB_TOKEN: ${HUGGINGFACEHUB_API_TOKEN}
      HF_TOKEN: ${HF_TOKEN}
      NEO4J_URI: ${NEO4J_URL}
      NEO4J_USERNAME: ${NEO4J_USERNAME}
      NEO4J_PASSWORD: ${NEO4J_PASSWORD}
      TGI_LLM_ENDPOINT: ${TGI_LLM_ENDPOINT}
      TEI_EMBEDDING_ENDPOINT: ${TEI_EMBEDDING_ENDPOINT}
      OPENAI_API_KEY: ${OPENAI_API_KEY}
      OPENAI_EMBEDDING_MODEL: ${OPENAI_EMBEDDING_MODEL}
      OPENAI_LLM_MODEL: ${OPENAI_LLM_MODEL}
      EMBEDDING_MODEL_ID: ${EMBEDDING_MODEL_ID}
      LLM_MODEL_ID: ${LLM_MODEL_ID}
      MAX_OUTPUT_TOKENS: ${MAX_OUTPUT_TOKENS}
      LOGFLAG: ${LOGFLAG}
      RETRIEVER_COMPONENT_NAME: "OPEA_RETRIEVER_NEO4J"
    restart: unless-stopped
  graphrag-gaudi-backend-server:
    image: ${REGISTRY:-opea}/graphrag:${TAG:-latest}
    container_name: graphrag-gaudi-backend-server
    depends_on:
      - neo4j-apoc
      - tei-embedding-service
      - retriever-neo4j-llamaindex
      - tgi-gaudi-service
    ports:
      - "8888:8888"
    environment:
      - no_proxy=${no_proxy}
      - https_proxy=${https_proxy}
      - http_proxy=${http_proxy}
      - MEGA_SERVICE_HOST_IP=graphrag-gaudi-backend-server
      - RETRIEVER_SERVICE_HOST_IP=retriever-neo4j-llamaindex
      - RETRIEVER_SERVICE_PORT=7000
      - LLM_SERVER_HOST_IP=tgi-gaudi-service
      - LLM_SERVER_PORT=${LLM_SERVER_PORT:-80}
      - LLM_MODEL_ID=${LLM_MODEL_ID}
      - LOGFLAG=${LOGFLAG}
    ipc: host
    restart: always
  graphrag-ui-server:
    image: ${REGISTRY:-opea}/graphrag-ui:${TAG:-latest}
    container_name: graphrag-ui-server
    depends_on:
      - graphrag-gaudi-backend-server
    ports:
      - "5173:5173"
    environment:
      - no_proxy=${no_proxy}
      - https_proxy=${https_proxy}
      - http_proxy=${http_proxy}
    ipc: host
    restart: always
  chatqna-gaudi-nginx-server:
    image: ${REGISTRY:-opea}/nginx:${TAG:-latest}
    container_name: chatqna-gaudi-nginx-server
    depends_on:
      - graphrag-gaudi-backend-server
      - graphrag-ui-server
    ports:
      - "${NGINX_PORT:-80}:80"
    environment:
      - no_proxy=${no_proxy}
      - https_proxy=${https_proxy}
      - http_proxy=${http_proxy}
      - FRONTEND_SERVICE_IP=graphrag-ui-server
      - FRONTEND_SERVICE_PORT=5173
      - BACKEND_SERVICE_NAME=graphrag
      - BACKEND_SERVICE_IP=graphrag-gaudi-backend-server
      - BACKEND_SERVICE_PORT=8888
      - DATAPREP_SERVICE_IP=dataprep-neo4j-llamaindex
      - DATAPREP_SERVICE_PORT=5000
    ipc: host
    restart: always
networks:
  default:
    driver: bridge