Add dataprep microservice to chatQnA example (#261)

* add dataprep microservice to chatQnA example, including the sampy yaml and test data.
Signed-off-by: zhlsunshine <huailong.zhang@intel.com>

microservices-connector/config/samples/chatQnA_dataprep_gaudi.yaml

@@ -0,0 +1,77 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+apiVersion: gmc.opea.io/v1alpha3
+kind: GMConnector
+metadata:
+  labels:
+    app.kubernetes.io/name: gmconnector
+    app.kubernetes.io/managed-by: kustomize
+    gmc/platform: gaudi
+  name: chatqa
+  namespace: chatqa
+spec:
+  routerConfig:
+    name: router
+    serviceName: router-service
+  nodes:
+    root:
+      routerType: Sequence
+      steps:
+      - name: Embedding
+        internalService:
+          serviceName: embedding-svc
+          config:
+            endpoint: /v1/embeddings
+            TEI_EMBEDDING_ENDPOINT: tei-embedding-gaudi-svc
+      - name: TeiEmbeddingGaudi
+        internalService:
+          serviceName: tei-embedding-gaudi-svc
+          isDownstreamService: true
+      - name: Retriever
+        data: $response
+        internalService:
+          serviceName: retriever-svc
+          config:
+            endpoint: /v1/retrieval
+            REDIS_URL: redis-vector-db
+            TEI_EMBEDDING_ENDPOINT: tei-embedding-gaudi-svc
+      - name: VectorDB
+        internalService:
+          serviceName: redis-vector-db
+          isDownstreamService: true
+      - name: Reranking
+        data: $response
+        internalService:
+          serviceName: reranking-svc
+          config:
+            endpoint: /v1/reranking
+            TEI_RERANKING_ENDPOINT: tei-reranking-svc
+      - name: TeiReranking
+        internalService:
+          serviceName: tei-reranking-svc
+          config:
+            endpoint: /rerank
+          isDownstreamService: true
+      - name: Llm
+        data: $response
+        internalService:
+          serviceName: llm-svc
+          config:
+            endpoint: /v1/chat/completions
+            TGI_LLM_ENDPOINT: tgi-gaudi-svc
+      - name: TgiGaudi
+        internalService:
+          serviceName: tgi-gaudi-svc
+          config:
+            endpoint: /generate
+          isDownstreamService: true
+      - name: DataPrep
+        internalService:
+          serviceName: data-prep-svc
+          config:
+            endpoint: /v1/dataprep
+            REDIS_URL: redis-vector-db
+            INDEX_NAME: data-prep
+            TEI_ENDPOINT: tei-embedding-svc
+          isDownstreamService: true

microservices-connector/config/samples/chatQnA_dataprep_xeon.yaml

@@ -0,0 +1,77 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+apiVersion: gmc.opea.io/v1alpha3
+kind: GMConnector
+metadata:
+  labels:
+    app.kubernetes.io/name: gmconnector
+    app.kubernetes.io/managed-by: kustomize
+    gmc/platform: xeon
+  name: chatqa
+  namespace: chatqa
+spec:
+  routerConfig:
+    name: router
+    serviceName: router-service
+  nodes:
+    root:
+      routerType: Sequence
+      steps:
+      - name: Embedding
+        internalService:
+          serviceName: embedding-svc
+          config:
+            endpoint: /v1/embeddings
+            TEI_EMBEDDING_ENDPOINT: tei-embedding-svc
+      - name: TeiEmbedding
+        internalService:
+          serviceName: tei-embedding-svc
+          isDownstreamService: true
+      - name: Retriever
+        data: $response
+        internalService:
+          serviceName: retriever-svc
+          config:
+            endpoint: /v1/retrieval
+            REDIS_URL: redis-vector-db
+            TEI_EMBEDDING_ENDPOINT: tei-embedding-svc
+      - name: VectorDB
+        internalService:
+          serviceName: redis-vector-db
+          isDownstreamService: true
+      - name: Reranking
+        data: $response
+        internalService:
+          serviceName: reranking-svc
+          config:
+            endpoint: /v1/reranking
+            TEI_RERANKING_ENDPOINT: tei-reranking-svc
+      - name: TeiReranking
+        internalService:
+          serviceName: tei-reranking-svc
+          config:
+            endpoint: /rerank
+          isDownstreamService: true
+      - name: Llm
+        data: $response
+        internalService:
+          serviceName: llm-svc
+          config:
+            endpoint: /v1/chat/completions
+            TGI_LLM_ENDPOINT: tgi-service-m
+      - name: Tgi
+        internalService:
+          serviceName: tgi-service-m
+          config:
+            endpoint: /generate
+          isDownstreamService: true
+      - name: DataPrep
+        internalService:
+          serviceName: data-prep-svc
+          config:
+            endpoint: /v1/dataprep
+            REDIS_URL: redis-vector-db
+            INDEX_NAME: data-prep
+            TEI_ENDPOINT: tei-embedding-svc
+          isDownstreamService: true

microservices-connector/test/data/gaudi.txt

@@ -0,0 +1,9 @@
+Intel® Gaudi® Al accelerators and Intel® Gaudi® software are designed to bring a new level of productivity advantages and choice to data center generative Al, offering high performance and efficiency for training neural networks. Here are some key features and information about Intel Gaudi:
+1. AI Training Processors: Gaudi processors are optimized for training deep learning models, with a focus on scalability and high performance.
+2. High Performance: Gaudi chips are designed to deliver high throughput and low latency for AI workloads. They leverage a high-bandwidth memory architecture and advanced interconnects to accelerate training processes.
+3. Scalability: Gaudi processors support scaling from single-node to large-scale, multi-node clusters. This makes them suitable for both small-scale and large-scale AI training environments.
+4. Efficient Architecture: The architecture of Gaudi processors is designed to maximize the utilization of compute resources, leading to efficient training of large models.
+5. Software Support: Gaudi processors come with software libraries and tools to facilitate integration with popular deep learning frameworks such as TensorFlow and PyTorch. This includes Habana's SynapseAI software stack, which provides optimized kernels and runtime for efficient execution on Gaudi hardware.
+6. Integration with Ecosystem: Gaudi processors can be integrated into various AI and cloud environments, making them versatile for different deployment scenarios.
+7. Competitive Alternative: Intel Gaudi provides an alternative to other AI processors such as NVIDIA GPUs, offering competitive performance and potentially better cost efficiency for certain workloads.
+Overall, Intel Gaudi processors are designed to provide high performance and efficiency for AI training, making them a valuable option for organizations looking to scale their deep learning capabilities.