Update docsum according to docker compose changes

.github/workflows/scripts/e2e/gmc_gaudi_test.sh

@@ -637,7 +637,7 @@ function validate_docsum() {
    export CLIENT_POD=$(kubectl get pod -n $DOCSUM_NAMESPACE -l app=client-test -o jsonpath={.items..metadata.name})
    echo "$CLIENT_POD"
    accessUrl=$(kubectl get gmc -n $DOCSUM_NAMESPACE -o jsonpath="{.items[?(@.metadata.name=='docsum')].status.accessUrl}")
-   kubectl exec "$CLIENT_POD" -n $DOCSUM_NAMESPACE -- curl $accessUrl  -X POST  -d '{"query":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}'  -H 'Content-Type: application/json' > $LOG_PATH/gmc_docsum.log
+   kubectl exec "$CLIENT_POD" -n $DOCSUM_NAMESPACE -- curl $accessUrl  -X POST  -d '{"type": "text", "messages":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}'  -H 'Content-Type: application/json' > $LOG_PATH/gmc_docsum.log
    exit_code=$?
    if [ $exit_code -ne 0 ]; then
        echo "docsum failed, please check the logs in ${LOG_PATH}!"

.github/workflows/scripts/e2e/gmc_xeon_test.sh

@@ -658,7 +658,7 @@ function validate_docsum() {
    export CLIENT_POD=$(kubectl get pod -n $DOCSUM_NAMESPACE -l app=client-test -o jsonpath={.items..metadata.name})
    echo "$CLIENT_POD"
    accessUrl=$(kubectl get gmc -n $DOCSUM_NAMESPACE -o jsonpath="{.items[?(@.metadata.name=='docsum')].status.accessUrl}")
-   kubectl exec "$CLIENT_POD" -n $DOCSUM_NAMESPACE -- curl $accessUrl  -X POST  -d '{"query":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}'  -H 'Content-Type: application/json' > $LOG_PATH/gmc_docsum.log
+   kubectl exec "$CLIENT_POD" -n $DOCSUM_NAMESPACE -- curl $accessUrl  -X POST  -d '{"type": "text", "messages":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}'  -H 'Content-Type: application/json' > $LOG_PATH/gmc_docsum.log
    exit_code=$?
    if [ $exit_code -ne 0 ]; then
        echo "docsum failed, please check the logs in ${LOG_PATH}!"

.github/workflows/scripts/e2e/manifest_gaudi_test.sh

@@ -86,7 +86,7 @@ function validate_docsum() {
     # Curl the DocSum LLM Service
     curl http://${ip_address}:${port}/v1/chat/docsum \
       -X POST \
-      -d '{"query":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}' \
+      -d '{"type": "text", "messages":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}' \
       -H 'Content-Type: application/json' > $LOG_PATH/curl_docsum.log
     exit_code=$?
     if [ $exit_code -ne 0 ]; then

.github/workflows/scripts/e2e/manifest_xeon_test.sh

@@ -86,7 +86,7 @@ function validate_docsum() {
     # Curl the DocSum LLM Service
     curl http://${ip_address}:${port}/v1/chat/docsum \
       -X POST \
-      -d '{"query":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}' \
+      -d '{"type", "text", "messages":"Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}' \
       -H 'Content-Type: application/json' > $LOG_PATH/curl_docsum.log
     exit_code=$?
     if [ $exit_code -ne 0 ]; then

helm-charts/common/ui/templates/configmap.yaml

@@ -16,6 +16,7 @@ data:
   BASE_URL: {{ .Values.BACKEND_SERVICE_ENDPOINT | quote }}
   {{- else if (contains "docsum-ui" .Values.image.repository) }}
   DOC_BASE_URL: {{ .Values.BACKEND_SERVICE_ENDPOINT | quote }}
+  BACKEND_SERVICE_ENDPOINT: {{ .Values.BACKEND_SERVICE_ENDPOINT | quote }}
   {{- else if (contains "docsum-react-ui" .Values.image.repository) }}
   VITE_DOC_SUM_URL: {{ .Values.BACKEND_SERVICE_ENDPOINT | quote }}
   {{- else if contains "chatqna-ui" .Values.image.repository }}

helm-charts/docsum/Chart.yaml

@@ -12,6 +12,9 @@ dependencies:
   - name: llm-uservice
     version: 1.0.0
     repository: "file://../common/llm-uservice"
+  - name: whisper
+    version: 1.0.0
+    repository: "file://../common/whisper"
   - name: ui
     version: 1.0.0
     repository: "file://../common/ui"

helm-charts/docsum/README.md

@@ -34,8 +34,10 @@ Open another terminal and run the following command to verify the service if wor
 
 ```console
 curl http://localhost:8888/v1/docsum \
-    -H 'Content-Type: application/json' \
-    -d '{"messages": "Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5."}'
+    -H 'Content-Type: multipart/form-data' \
+    -F "type=text" \
+    -F "messages=Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5." \
+    -F "max_tokens=32"
 ```
 
 ### Verify the workload through UI

helm-charts/docsum/templates/deployment.yaml

@@ -39,8 +39,8 @@ spec:
               {{- else }}
               value: {{ .Release.Name }}-llm-uservice
               {{- end }}
-            #- name: MEGA_SERVICE_PORT
-            #  value: {{ .Values.port }}
+            - name: DATA_SERVICE_HOST_IP
+              value: {{ .Release.Name }}-m2t
           securityContext:
             {{- toYaml .Values.securityContext | nindent 12 }}
           image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"

helm-charts/docsum/templates/m2t.yaml

@@ -0,0 +1,94 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: {{ .Release.Name }}-m2t
+  labels:
+    {{- include "docsum.labels" . | nindent 4 }}
+    app: {{ .Release.Name }}-m2t
+spec:
+  replicas: {{ .Values.replicaCount }}
+  selector:
+    matchLabels:
+      {{- include "docsum.selectorLabels" . | nindent 6 }}
+      app: {{ .Release.Name }}-m2t
+  template:
+    metadata:
+      {{- with .Values.podAnnotations }}
+      annotations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      labels:
+        {{- include "docsum.selectorLabels" . | nindent 8 }}
+        app: {{ .Release.Name }}-m2t
+    spec:
+      {{- with .Values.imagePullSecrets }}
+      imagePullSecrets:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      securityContext:
+        {{- toYaml .Values.podSecurityContext | nindent 8 }}
+      containers:
+        - name: {{ .Release.Name }}-m2t
+          env:
+            - name: V2A_ENDPOINT
+              value: {{ .Release.Name }}-v2a:{{ .Values.v2a.service.port }}
+            - name: A2T_ENDPOINT
+              value: {{ .Release.Name }}-whisper:{{ .Values.whisper.service.port }}
+          securityContext:
+            {{- toYaml .Values.securityContext | nindent 12 }}
+          image: "{{ .Values.m2t.image.repository }}:{{ .Values.m2t.image.tag | default .Chart.AppVersion }}"
+          imagePullPolicy: {{ .Values.image.pullPolicy }}
+          volumeMounts:
+            - mountPath: /tmp
+              name: tmp
+          ports:
+            - name: m2t
+              containerPort: {{ .Values.m2t.port }}
+              protocol: TCP
+          resources:
+            {{- toYaml .Values.resources | nindent 12 }}
+      volumes:
+        - name: tmp
+          emptyDir: {}
+      {{- with .Values.nodeSelector }}
+      nodeSelector:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.affinity }}
+      affinity:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.tolerations }}
+      tolerations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- if .Values.evenly_distributed }}
+      topologySpreadConstraints:
+        - maxSkew: 1
+          topologyKey: kubernetes.io/hostname
+          whenUnsatisfiable: ScheduleAnyway
+          labelSelector:
+            matchLabels:
+              {{- include "docsum.selectorLabels" . | nindent 14 }}
+              app: {{ .Release.Name }}-m2t
+      {{- end }}
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ .Release.Name }}-m2t
+  labels:
+    {{- include "docsum.labels" . | nindent 4 }}
+spec:
+  type: {{ .Values.m2t.service.type }}
+  ports:
+    - port: {{ .Values.m2t.service.port }}
+      targetPort: {{ .Values.m2t.port }}
+      protocol: TCP
+      name: m2t
+  selector:
+    {{- include "docsum.selectorLabels" . | nindent 4 }}
+    app: {{ .Release.Name }}-m2t

helm-charts/docsum/templates/tests/test-pod.yaml

@@ -21,6 +21,7 @@ spec:
           for ((i=1; i<=max_retry; i++)); do
             curl http://{{ include "docsum.fullname" . }}:{{ .Values.service.port }}/v1/docsum -sS --fail-with-body \
             -H 'Content-Type: multipart/form-data' \
+            -H "type=text" \
             -F "messages=Text Embeddings Inference (TEI) is a toolkit for deploying and serving open source text embeddings and sequence classification models. TEI enables high-performance extraction for the most popular models, including FlagEmbedding, Ember, GTE and E5." \
             -F "max_tokens=32" && break;
             curlcode=$?

helm-charts/docsum/templates/v2a.yaml

@@ -0,0 +1,89 @@
+# Copyright (C) 2024 Intel Corporation
+# SPDX-License-Identifier: Apache-2.0
+
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: {{ .Release.Name }}-v2a
+  labels:
+    {{- include "docsum.labels" . | nindent 4 }}
+    app: {{ .Release.Name }}-v2a
+spec:
+  replicas: {{ .Values.replicaCount }}
+  selector:
+    matchLabels:
+      {{- include "docsum.selectorLabels" . | nindent 6 }}
+      app: {{ .Release.Name }}-v2a
+  template:
+    metadata:
+      {{- with .Values.podAnnotations }}
+      annotations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      labels:
+        {{- include "docsum.selectorLabels" . | nindent 8 }}
+        app: {{ .Release.Name }}-v2a
+    spec:
+      {{- with .Values.imagePullSecrets }}
+      imagePullSecrets:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      securityContext:
+        {{- toYaml .Values.podSecurityContext | nindent 8 }}
+      containers:
+        - name: {{ .Release.Name }}-v2a
+          securityContext:
+            {{- toYaml .Values.securityContext | nindent 12 }}
+          image: "{{ .Values.v2a.image.repository }}:{{ .Values.v2a.image.tag | default .Chart.AppVersion }}"
+          imagePullPolicy: {{ .Values.image.pullPolicy }}
+          volumeMounts:
+            - mountPath: /tmp
+              name: tmp
+          ports:
+            - name: v2a
+              containerPort: {{ .Values.v2a.port }}
+              protocol: TCP
+          resources:
+            {{- toYaml .Values.resources | nindent 12 }}
+      volumes:
+        - name: tmp
+          emptyDir: {}
+      {{- with .Values.nodeSelector }}
+      nodeSelector:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.affinity }}
+      affinity:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- with .Values.tolerations }}
+      tolerations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}
+      {{- if .Values.evenly_distributed }}
+      topologySpreadConstraints:
+        - maxSkew: 1
+          topologyKey: kubernetes.io/hostname
+          whenUnsatisfiable: ScheduleAnyway
+          labelSelector:
+            matchLabels:
+              {{- include "docsum.selectorLabels" . | nindent 14 }}
+              app: {{ .Release.Name }}-v2a
+      {{- end }}
+---
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ .Release.Name }}-v2a
+  labels:
+    {{- include "docsum.labels" . | nindent 4 }}
+spec:
+  type: {{ .Values.v2a.service.type }}
+  ports:
+    - port: {{ .Values.v2a.service.port }}
+      targetPort: {{ .Values.v2a.port }}
+      protocol: TCP
+      name: v2a
+  selector:
+    {{- include "docsum.selectorLabels" . | nindent 4 }}
+    app: {{ .Release.Name }}-v2a

helm-charts/docsum/values.yaml

@@ -13,6 +13,24 @@ image:
   pullPolicy: IfNotPresent
   # Overrides the image tag whose default is the chart appVersion.
   tag: "latest"
+v2a:
+  image:
+    repository: opea/dataprep-video2audio
+    # Overrides the image tag whose default is the chart appVersion.
+    tag: "latest"
+  port: 7078
+  service:
+    type: ClusterIP
+    port: 7078
+m2t:
+  image:
+    repository: opea/dataprep-multimedia2text
+    # Overrides the image tag whose default is the chart appVersion.
+    tag: "latest"
+  port: 7079
+  service:
+    type: ClusterIP
+    port: 7079
 
 port: 8888
 service:

microservices-connector/config/manifests/docsum_svelte-ui.yaml

@@ -15,6 +15,7 @@ metadata:
     app.kubernetes.io/managed-by: Helm
 data:
   DOC_BASE_URL: "/v1/docsum"
+  BACKEND_SERVICE_ENDPOINT: "/v1/docsum"
 ---
 # Source: ui/templates/service.yaml
 # Copyright (C) 2024 Intel Corporation