From e80e567817439af1b70b56ff4a60fa58c24e2439 Mon Sep 17 00:00:00 2001 From: Ke Ding Date: Fri, 21 Jun 2024 07:24:18 -0700 Subject: [PATCH] Add gpu support for ChatQnA (#308) * add gpu support for chatqna Signed-off-by: Ding, Ke * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --------- Signed-off-by: Ding, Ke Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> --- ChatQnA/README.md | 4 + ChatQnA/docker/gpu/README.md | 253 +++++++++++++++++++++++++ ChatQnA/docker/gpu/docker_compose.yaml | 212 +++++++++++++++++++++ README.md | 2 +- 4 files changed, 470 insertions(+), 1 deletion(-) create mode 100644 ChatQnA/docker/gpu/README.md create mode 100644 ChatQnA/docker/gpu/docker_compose.yaml diff --git a/ChatQnA/README.md b/ChatQnA/README.md index 740ea890c..42cac30b1 100644 --- a/ChatQnA/README.md +++ b/ChatQnA/README.md @@ -22,6 +22,10 @@ Refer to the [Gaudi Guide](./docker/gaudi/README.md) for instructions on deployi Refer to the [Xeon Guide](./docker/xeon/README.md) for instructions on deploying ChatQnA on Xeon. +## Deploy ChatQnA on NVIDIA GPU + +Refer to the [NVIDIA GPU Guide](./docker/gpu/README.md) for instructions on deploying ChatQnA on NVIDIA GPU. + ## Deploy ChatQnA into Kubernetes on Xeon & Gaudi Refer to the [Kubernetes Guide](./kubernetes/manifests/README.md) for instructions on deploying ChatQnA into Kubernetes on Xeon & Gaudi. diff --git a/ChatQnA/docker/gpu/README.md b/ChatQnA/docker/gpu/README.md new file mode 100644 index 000000000..18b3e9572 --- /dev/null +++ b/ChatQnA/docker/gpu/README.md @@ -0,0 +1,253 @@ +# Build MegaService of ChatQnA on NVIDIA GPU + +This document outlines the deployment process for a ChatQnA application utilizing the [GenAIComps](https://github.com/opea-project/GenAIComps.git) microservice pipeline on NVIDIA GPU platform. The steps include Docker image creation, container deployment via Docker Compose, and service execution to integrate microservices such as embedding, retriever, rerank, and llm. We will publish the Docker images to Docker Hub, it will simplify the deployment process for this service. + +## 🚀 Build Docker Images + +First of all, you need to build Docker Images locally. This step can be ignored after the Docker images published to Docker hub. + +### 1. Source Code install GenAIComps + +```bash +git clone https://github.com/opea-project/GenAIComps.git +cd GenAIComps +``` + +### 2. Build Embedding Image + +```bash +docker build --no-cache -t opea/embedding-tei:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f comps/embeddings/langchain/docker/Dockerfile . +``` + +### 3. Build Retriever Image + +```bash +docker build --no-cache -t opea/retriever-redis:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f comps/retrievers/langchain/redis/docker/Dockerfile . +``` + +### 4. Build Rerank Image + +```bash +docker build --no-cache -t opea/reranking-tei:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f comps/reranks/langchain/docker/Dockerfile . +``` + +### 5. Build LLM Image + +```bash +docker build --no-cache -t opea/llm-tgi:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f comps/llms/text-generation/tgi/Dockerfile . +``` + +### 6. Build Dataprep Image + +```bash +docker build --no-cache -t opea/dataprep-redis:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f comps/dataprep/redis/langchain/docker/Dockerfile . +``` + +### 7. Build MegaService Docker Image + +To construct the Mega Service, we utilize the [GenAIComps](https://github.com/opea-project/GenAIComps.git) microservice pipeline within the `chatqna.py` Python script. Build the MegaService Docker image using the command below: + +```bash +git clone https://github.com/opea-project/GenAIExamples.git +cd GenAIExamples/ChatQnA/docker +docker build --no-cache -t opea/chatqna:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f Dockerfile . +cd ../../.. +``` + +### 8. Build UI Docker Image + +Construct the frontend Docker image using the command below: + +```bash +cd GenAIExamples/ChatQnA/docker/ui/ +docker build --no-cache -t opea/chatqna-ui:latest --build-arg https_proxy=$https_proxy --build-arg http_proxy=$http_proxy -f ./docker/Dockerfile . +cd ../../../.. +``` + +Then run the command `docker images`, you will have the following 7 Docker Images: + +1. `opea/embedding-tei:latest` +2. `opea/retriever-redis:latest` +3. `opea/reranking-tei:latest` +4. `opea/llm-tgi:latest` +5. `opea/dataprep-redis:latest` +6. `opea/chatqna:latest` +7. `opea/chatqna-ui:latest` + +## 🚀 Start MicroServices and MegaService + +### Setup Environment Variables + +Since the `docker_compose.yaml` will consume some environment variables, you need to setup them in advance as below. + +```bash +export no_proxy=${your_no_proxy} +export http_proxy=${your_http_proxy} +export https_proxy=${your_http_proxy} +export EMBEDDING_MODEL_ID="BAAI/bge-base-en-v1.5" +export RERANK_MODEL_ID="BAAI/bge-reranker-base" +export LLM_MODEL_ID="Intel/neural-chat-7b-v3-3" +export TEI_EMBEDDING_ENDPOINT="http://${host_ip}:8090" +export TEI_RERANKING_ENDPOINT="http://${host_ip}:8808" +export TGI_LLM_ENDPOINT="http://${host_ip}:8008" +export REDIS_URL="redis://${host_ip}:6379" +export INDEX_NAME="rag-redis" +export HUGGINGFACEHUB_API_TOKEN=${your_hf_api_token} +export MEGA_SERVICE_HOST_IP=${host_ip} +export EMBEDDING_SERVICE_HOST_IP=${host_ip} +export RETRIEVER_SERVICE_HOST_IP=${host_ip} +export RERANK_SERVICE_HOST_IP=${host_ip} +export LLM_SERVICE_HOST_IP=${host_ip} +export BACKEND_SERVICE_ENDPOINT="http://${host_ip}:8888/v1/chatqna" +export DATAPREP_SERVICE_ENDPOINT="http://${host_ip}:6007/v1/dataprep" +``` + +Note: Please replace with `host_ip` with you external IP address, do **NOT** use localhost. + +### Start all the services Docker Containers + +```bash +cd GenAIExamples/ChatQnA/docker/gpu/ +docker compose -f docker_compose.yaml up -d +``` + +### Validate MicroServices and MegaService + +1. TEI Embedding Service + +```bash +curl ${host_ip}:8090/embed \ + -X POST \ + -d '{"inputs":"What is Deep Learning?"}' \ + -H 'Content-Type: application/json' +``` + +2. Embedding Microservice + +```bash +curl http://${host_ip}:6000/v1/embeddings \ + -X POST \ + -d '{"text":"hello"}' \ + -H 'Content-Type: application/json' +``` + +3. Retriever Microservice + +To consume the retriever microservice, you need to generate a mock embedding vector of length 768 in Python script: + +```python +import random + +embedding = [random.uniform(-1, 1) for _ in range(768)] +print(embedding) +``` + +Then substitute your mock embedding vector for the `${your_embedding}` in the following `curl` command: + +```bash +curl http://${host_ip}:7000/v1/retrieval \ + -X POST \ + -d '{"text":"test", "embedding":${your_embedding}}' \ + -H 'Content-Type: application/json' +``` + +4. TEI Reranking Service + +```bash +curl http://${host_ip}:8808/rerank \ + -X POST \ + -d '{"query":"What is Deep Learning?", "texts": ["Deep Learning is not...", "Deep learning is..."]}' \ + -H 'Content-Type: application/json' +``` + +5. Reranking Microservice + +```bash +curl http://${host_ip}:8000/v1/reranking \ + -X POST \ + -d '{"initial_query":"What is Deep Learning?", "retrieved_docs": [{"text":"Deep Learning is not..."}, {"text":"Deep learning is..."}]}' \ + -H 'Content-Type: application/json' +``` + +6. TGI Service + +```bash +curl http://${host_ip}:8008/generate \ + -X POST \ + -d '{"inputs":"What is Deep Learning?","parameters":{"max_new_tokens":64, "do_sample": true}}' \ + -H 'Content-Type: application/json' +``` + +7. LLM Microservice + +```bash +curl http://${host_ip}:9000/v1/chat/completions \ + -X POST \ + -d '{"query":"What is Deep Learning?","max_new_tokens":17,"top_k":10,"top_p":0.95,"typical_p":0.95,"temperature":0.01,"repetition_penalty":1.03,"streaming":true}' \ + -H 'Content-Type: application/json' +``` + +8. MegaService + +```bash +curl http://${host_ip}:8888/v1/chatqna -H "Content-Type: application/json" -d '{ + "messages": "What is the revenue of Nike in 2023?" + }' +``` + +9. Dataprep Microservice（Optional） + +If you want to update the default knowledge base, you can use the following commands: + +Update Knowledge Base via Local File Upload: + +```bash +curl -X POST "http://${host_ip}:6007/v1/dataprep" \ + -H "Content-Type: multipart/form-data" \ + -F "files=@./nke-10k-2023.pdf" +``` + +This command updates a knowledge base by uploading a local file for processing. Update the file path according to your environment. + +Add Knowledge Base via HTTP Links: + +```bash +curl -X POST "http://${host_ip}:6007/v1/dataprep" \ + -H "Content-Type: multipart/form-data" \ + -F 'link_list=["https://opea.dev"]' +``` + +This command updates a knowledge base by submitting a list of HTTP links for processing. + +## Enable LangSmith for Monotoring Application (Optional) + +LangSmith offers tools to debug, evaluate, and monitor language models and intelligent agents. It can be used to assess benchmark data for each microservice. Before launching your services with `docker compose -f docker_compose.yaml up -d`, you need to enable LangSmith tracing by setting the `LANGCHAIN_TRACING_V2` environment variable to true and configuring your LangChain API key. + +Here's how you can do it: + +1. Install the latest version of LangSmith: + +```bash +pip install -U langsmith +``` + +2. Set the necessary environment variables: + +```bash +export LANGCHAIN_TRACING_V2=true +export LANGCHAIN_API_KEY=ls_... +``` + +## 🚀 Launch the UI + +To access the frontend, open the following URL in your browser: http://{host_ip}:5173. By default, the UI runs on port 5173 internally. If you prefer to use a different host port to access the frontend, you can modify the port mapping in the `docker_compose.yaml` file as shown below: + +```yaml + chaqna-ui-server: + image: opea/chatqna-ui:latest + ... + ports: + - "80:5173" +``` + +![project-screenshot](../../assets/img/chat_ui_init.png) diff --git a/ChatQnA/docker/gpu/docker_compose.yaml b/ChatQnA/docker/gpu/docker_compose.yaml new file mode 100644 index 000000000..e19fcdb86 --- /dev/null +++ b/ChatQnA/docker/gpu/docker_compose.yaml @@ -0,0 +1,212 @@ + +# Copyright (C) 2024 Intel Corporation +# SPDX-License-Identifier: Apache-2.0 + +version: "3.8" + +services: + redis-vector-db: + image: redis/redis-stack:7.2.0-v9 + container_name: redis-vector-db + ports: + - "6379:6379" + - "8001:8001" + dataprep-redis-service: + image: opea/dataprep-redis:latest + container_name: dataprep-redis-server + depends_on: + - redis-vector-db + ports: + - "6007:6007" + environment: + no_proxy: ${no_proxy} + http_proxy: ${http_proxy} + https_proxy: ${https_proxy} + REDIS_URL: ${REDIS_URL} + INDEX_NAME: ${INDEX_NAME} + tei-embedding-service: + image: ghcr.io/huggingface/text-embeddings-inference:1.2 + container_name: tei-embedding-server + ports: + - "8090:80" + volumes: + - "./data:/data" + shm_size: 1g + ipc: host + environment: + no_proxy: ${no_proxy} + http_proxy: ${http_proxy} + https_proxy: ${https_proxy} + command: --model-id ${EMBEDDING_MODEL_ID} + deploy: + resources: + reservations: + devices: + - driver: nvidia + count: 1 + capabilities: [gpu] + embedding: + image: opea/embedding-tei:latest + container_name: embedding-tei-server + depends_on: + - tei-embedding-service + ports: + - "6000:6000" + ipc: host + environment: + no_proxy: ${no_proxy} + http_proxy: ${http_proxy} + https_proxy: ${https_proxy} + TEI_EMBEDDING_ENDPOINT: ${TEI_EMBEDDING_ENDPOINT} + LANGCHAIN_API_KEY: ${LANGCHAIN_API_KEY} + LANGCHAIN_TRACING_V2: ${LANGCHAIN_TRACING_V2} + LANGCHAIN_PROJECT: "opea-embedding-service" + restart: unless-stopped + retriever: + image: opea/retriever-redis:latest + container_name: retriever-redis-server + depends_on: + - redis-vector-db + ports: + - "7000:7000" + ipc: host + environment: + no_proxy: ${no_proxy} + http_proxy: ${http_proxy} + https_proxy: ${https_proxy} + REDIS_URL: ${REDIS_URL} + INDEX_NAME: ${INDEX_NAME} + LANGCHAIN_API_KEY: ${LANGCHAIN_API_KEY} + LANGCHAIN_TRACING_V2: ${LANGCHAIN_TRACING_V2} + LANGCHAIN_PROJECT: "opea-retriever-service" + restart: unless-stopped + tei-reranking-service: + image: ghcr.io/huggingface/text-embeddings-inference:1.2 + container_name: tei-reranking-server + ports: + - "8808:80" + volumes: + - "./data:/data" + shm_size: 1g + environment: + no_proxy: ${no_proxy} + http_proxy: ${http_proxy} + https_proxy: ${https_proxy} + HUGGINGFACEHUB_API_TOKEN: ${HUGGINGFACEHUB_API_TOKEN} + HF_HUB_DISABLE_PROGRESS_BARS: 1 + HF_HUB_ENABLE_HF_TRANSFER: 0 + command: --model-id ${RERANK_MODEL_ID} + deploy: + resources: + reservations: + devices: + - driver: nvidia + count: 1 + capabilities: [gpu] + reranking: + image: opea/reranking-tei:latest + container_name: reranking-tei-server + depends_on: + - tei-reranking-service + ports: + - "8000:8000" + ipc: host + environment: + no_proxy: ${no_proxy} + http_proxy: ${http_proxy} + https_proxy: ${https_proxy} + TEI_RERANKING_ENDPOINT: ${TEI_RERANKING_ENDPOINT} + HUGGINGFACEHUB_API_TOKEN: ${HUGGINGFACEHUB_API_TOKEN} + HF_HUB_DISABLE_PROGRESS_BARS: 1 + HF_HUB_ENABLE_HF_TRANSFER: 0 + LANGCHAIN_API_KEY: ${LANGCHAIN_API_KEY} + LANGCHAIN_TRACING_V2: ${LANGCHAIN_TRACING_V2} + LANGCHAIN_PROJECT: "opea-reranking-service" + restart: unless-stopped + tgi-service: + image: ghcr.io/huggingface/text-generation-inference:2.0 + container_name: tgi-server + ports: + - "8008:80" + volumes: + - "./data:/data" + environment: + no_proxy: ${no_proxy} + http_proxy: ${http_proxy} + https_proxy: ${https_proxy} + HF_TOKEN: ${HUGGINGFACEHUB_API_TOKEN} + HF_HUB_DISABLE_PROGRESS_BARS: 1 + HF_HUB_ENABLE_HF_TRANSFER: 0 + ipc: host + command: --model-id ${LLM_MODEL_ID} + deploy: + resources: + reservations: + devices: + - driver: nvidia + count: 1 + capabilities: [gpu] + llm: + image: opea/llm-tgi:latest + container_name: llm-tgi-server + depends_on: + - tgi-service + ports: + - "9000:9000" + ipc: host + environment: + no_proxy: ${no_proxy} + http_proxy: ${http_proxy} + https_proxy: ${https_proxy} + TGI_LLM_ENDPOINT: ${TGI_LLM_ENDPOINT} + HUGGINGFACEHUB_API_TOKEN: ${HUGGINGFACEHUB_API_TOKEN} + HF_HUB_DISABLE_PROGRESS_BARS: 1 + HF_HUB_ENABLE_HF_TRANSFER: 0 + LANGCHAIN_API_KEY: ${LANGCHAIN_API_KEY} + LANGCHAIN_TRACING_V2: ${LANGCHAIN_TRACING_V2} + LANGCHAIN_PROJECT: "opea-llm-service" + restart: unless-stopped + chaqna-backend-server: + image: opea/chatqna:latest + container_name: chatqna-backend-server + depends_on: + - redis-vector-db + - tei-embedding-service + - embedding + - retriever + - tei-reranking-service + - reranking + - tgi-service + - llm + ports: + - "8888:8888" + environment: + - no_proxy=${no_proxy} + - https_proxy=${https_proxy} + - http_proxy=${http_proxy} + - MEGA_SERVICE_HOST_IP=${MEGA_SERVICE_HOST_IP} + - EMBEDDING_SERVICE_HOST_IP=${EMBEDDING_SERVICE_HOST_IP} + - RETRIEVER_SERVICE_HOST_IP=${RETRIEVER_SERVICE_HOST_IP} + - RERANK_SERVICE_HOST_IP=${RERANK_SERVICE_HOST_IP} + - LLM_SERVICE_HOST_IP=${LLM_SERVICE_HOST_IP} + ipc: host + restart: always + chaqna-ui-server: + image: opea/chatqna-ui:latest + container_name: chatqna-ui-server + depends_on: + - chaqna-backend-server + ports: + - "5173:5173" + environment: + - no_proxy=${no_proxy} + - https_proxy=${https_proxy} + - http_proxy=${http_proxy} + - CHAT_BASE_URL=${BACKEND_SERVICE_ENDPOINT} + - UPLOAD_FILE_BASE_URL=${DATAPREP_SERVICE_ENDPOINT} + ipc: host + restart: always + +networks: + default: + driver: bridge diff --git a/README.md b/README.md index c2cb68c96..0a2629f49 100644 --- a/README.md +++ b/README.md @@ -45,7 +45,7 @@ All the examples are well-validated on Intel platforms. In addition, these examp BGE-Base Redis TGI TEI - Xeon/Gaudi2 + Xeon/Gaudi2/GPU Chatbot