release v1.1.0-alpha0 (#459)

* add rag kuberay and jupyterhub image (#440) * Rollback to previous image (#454) * Ray Webhook Support for Single-Host, Multi-Slice TPUs (#453) * Fix incorrect replicaIndex for single-host, multi replica * Fix single-host, multi-slice deletion logic * Update README & simplify workloads.tfvars for RAG (#445) * RAG marketplace updates (#456) * fix RAG marketplace changes --------- Co-authored-by: Chia-Yi Liang <chiayiliang327@gmail.com> Co-authored-by: zlq <zlq@google.com> Co-authored-by: ryanaoleary <113500783+ryanaoleary@users.noreply.github.com> Co-authored-by: imreddy13 <132504814+imreddy13@users.noreply.github.com>
GoogleCloudPlatform · Mar 27, 2024 · a3560f4 · a3560f4
1 parent 515828e
commit a3560f4
Show file tree

Hide file tree

Showing 22 changed files with 404 additions and 450 deletions.
diff --git a/applications/rag/README.md b/applications/rag/README.md
diff --git a/applications/rag/example_notebooks/rag-kaggle-ray-sql-latest.ipynb b/applications/rag/example_notebooks/rag-kaggle-ray-sql-latest.ipynb
@@ -20,8 +20,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!pip install langchain==0.1.9 ray==2.9.3 datasets==2.18.0 sentence-transformers==2.5.1 kaggle==1.6.6\n",
-    "!pip install \"cloud-sql-python-connector[pg8000]==1.7.0\" SQLAlchemy==2.0.7"
+    "!pip install ray==2.9.3 kaggle==1.6.6"
    ]
   },
   {
@@ -122,7 +121,7 @@
     "BATCH_SIZE = 100\n",
     "CHUNK_SIZE = 1000 # text chunk sizes which will be converted to vector embeddings\n",
     "CHUNK_OVERLAP = 10\n",
-    "TABLE_NAME = 'googlemaps_reviews_db'  # CloudSQL table name\n",
+    "TABLE_NAME = 'netflix_reviews_db'  # CloudSQL table name\n",
     "DIMENSION = 384  # Embeddings size\n",
     "ACTOR_POOL_SIZE = 1 # number of actors for the distributed map_batches function\n",
     "\n",
@@ -272,6 +271,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# Port forward to the Ray dashboard and go to `localhost:8265` in a browser to see job status: kubectl port-forward -n <namespace> service/ray-cluster-kuberay-head-svc 8265:8265\n",
     "import time\n",
     "\n",
     "start_time = time.time()\n",

diff --git a/applications/rag/frontend/main.tf b/applications/rag/frontend/main.tf
@@ -109,7 +109,7 @@ resource "kubernetes_deployment" "rag_frontend_deployment" {
       spec {
         service_account_name = var.google_service_account
         container {
-          image = "us-central1-docker.pkg.dev/ai-on-gke/rag-on-gke/frontend@sha256:91835a736252a43adfee6aa552a4853442d3c8b1dd9e5a82383cf25f536af6d0"
+          image = "us-central1-docker.pkg.dev/ai-on-gke/rag-on-gke/frontend@sha256:7530314f46a7c4f487240984f6412d4041311a3252238ac3635375434521f837"
           name  = "rag-frontend"
 
           port {

diff --git a/applications/rag/main.tf b/applications/rag/main.tf
@@ -228,6 +228,7 @@ module "kuberay-cluster" {
   grafana_host           = module.kuberay-monitoring.grafana_uri
   disable_network_policy = var.disable_ray_cluster_network_policy
   depends_on             = [module.kuberay-operator]
+  use_custom_image       = true
 
   # IAP Auth parameters
   add_auth                 = var.ray_dashboard_add_auth

diff --git a/applications/rag/metadata.display.yaml b/applications/rag/metadata.display.yaml
@@ -16,11 +16,11 @@ spec:
       variables:
         acknowledge:
           name: acknowledge
-          title: Confirm that all prerequisites have been met. 
+          title: Confirm that all prerequisites have been met.
           section: acknowledge
           subtext: If you choose an existing GKE standard cluster, ensure it has an L4 GPU Node pool with <a href="https://cloud.google.com/kubernetes-engine/docs/how-to/cluster-autoscaler#enable_autoscaling">autoscaling</a> and <a href="https://cloud.google.com/kubernetes-engine/docs/how-to/node-auto-provisioning#enable">node auto-provisioning</a> enabled.
           enumValueLabels:
-            - label: "Confirm that all prerequisites have been met."
+            - label: Confirm that all prerequisites have been met.
               value: "true"
         additional_labels:
           name: additional_labels
@@ -36,6 +36,7 @@ spec:
               value: "true"
             - label: Standard Cluster
               value: "false"
+          invisible: true
         cloudsql_instance:
           name: cloudsql_instance
           title: CloudSQL Instance
@@ -121,50 +122,10 @@ spec:
           name: frontend_add_auth
           title: Enable IAP for Frontend Application
           section: rag_iap_auth
-        frontend_client_id:
-          name: frontend_client_id
-          title: Frontend Client Id
-          invisible: true
-          section: rag_iap_auth
-        frontend_client_secret:
-          name: frontend_client_secret
-          title: Frontend Client Secret
-          invisible: true
-          section: rag_iap_auth
         frontend_domain:
           name: frontend_domain
           title: Frontend Domain
           section: rag_iap_auth
-        frontend_k8s_backend_config_name:
-          name: frontend_k8s_backend_config_name
-          title: Frontend K8s Backend Config Name
-          invisible: true
-          section: rag_iap_auth
-        frontend_k8s_backend_service_name:
-          name: frontend_k8s_backend_service_name
-          title: Frontend K8s Backend Service Name
-          invisible: true
-          section: rag_iap_auth
-        frontend_k8s_backend_service_port:
-          name: frontend_k8s_backend_service_port
-          title: Frontend K8s Backend Service Port
-          invisible: true
-          section: rag_iap_auth
-        frontend_k8s_iap_secret_name:
-          name: frontend_k8s_iap_secret_name
-          title: Frontend K8s Iap Secret Name
-          invisible: true
-          section: rag_iap_auth
-        frontend_k8s_ingress_name:
-          name: frontend_k8s_ingress_name
-          title: Frontend K8s Ingress Name
-          invisible: true
-          section: rag_iap_auth
-        frontend_k8s_managed_cert_name:
-          name: frontend_k8s_managed_cert_name
-          title: Frontend K8s Managed Cert Name
-          invisible: true
-          section: rag_iap_auth
         frontend_members_allowlist:
           name: frontend_members_allowlist
           title: Frontend Members Allowlist
@@ -178,105 +139,14 @@ spec:
         goog_cm_deployment_name:
           name: goog_cm_deployment_name
           title: Goog Cm Deployment Name
-        gpu_pools:
-          name: gpu_pools
-          title: Gpu Pools
-          invisible: true
-        ray_dashboard_add_auth:
-          name: ray_dashboard_add_auth
-          title: Enable IAP for Ray Dashboard Application
-          section: ray_iap_auth
-        ray_dashboard_client_id:
-          name: ray_dashboard_client_id
-          title: Ray Dashboard Client Id
-          section: ray_iap_auth
-          invisible: true
-        ray_dashboard_client_secret:
-          name: ray_dashboard_client_secret
-          title: Ray Dashboard Client Secret
-          section: ray_iap_auth
-          invisible: true
-        ray_dashboard_domain:
-          name: ray_dashboard_domain
-          title: Ray Dashboard Domain
-          section: ray_iap_auth
-        ray_dashboard_k8s_backend_config_name:
-          name: ray_dashboard_k8s_backend_config_name
-          title: Ray Dashboard K8s Backend Config Name
-          section: ray_iap_auth
-          invisible: true
-        ray_dashboard_k8s_backend_service_port:
-          name: ray_dashboard_k8s_backend_service_port
-          title: Ray Dashboard K8s Backend Service Port
-          section: ray_iap_auth
-          invisible: true
-        ray_dashboard_k8s_iap_secret_name:
-          name: ray_dashboard_k8s_iap_secret_name
-          title: Ray Dashboard K8s Iap Secret Name
-          section: ray_iap_auth
-          invisible: true
-        ray_dashboard_k8s_ingress_name:
-          name: ray_dashboard_k8s_ingress_name
-          title: Ray Dashboard K8s Ingress Name
-          section: ray_iap_auth
-          invisible: true
-        ray_dashboard_k8s_managed_cert_name:
-          name: ray_dashboard_k8s_managed_cert_name
-          title: Ray Dashboard K8s Managed Cert Name
-          section: ray_iap_auth
-          invisible: true
-        ray_dashboard_members_allowlist:
-          name: ray_dashboard_members_allowlist
-          title: Ray Dashboard Members Allowlist
-          section: ray_iap_auth
         jupyter_add_auth:
           name: jupyter_add_auth
           title: Enable IAP for JupyterHub Application
           section: jupyter_iap_auth
-        jupyter_client_id:
-          name: jupyter_client_id
-          title: Jupyter Client Id
-          invisible: true
-          section: jupyter_iap_auth
-        jupyter_client_secret:
-          name: jupyter_client_secret
-          title: Jupyter Client Secret
-          invisible: true
-          section: jupyter_iap_auth
         jupyter_domain:
           name: jupyter_domain
           title: Jupyter Domain
           section: jupyter_iap_auth
-        jupyter_k8s_backend_config_name:
-          name: jupyter_k8s_backend_config_name
-          title: Jupyter K8s Backend Config Name
-          invisible: true
-          section: jupyter_iap_auth
-        jupyter_k8s_backend_service_name:
-          name: jupyter_k8s_backend_service_name
-          title: Jupyter K8s Backend Service Name
-          invisible: true
-          section: jupyter_iap_auth
-        jupyter_k8s_backend_service_port:
-          name: jupyter_k8s_backend_service_port
-          title: Jupyter K8s Backend Service Port
-          invisible: true
-          section: jupyter_iap_auth
-        jupyter_k8s_iap_secret_name:
-          name: jupyter_k8s_iap_secret_name
-          title: Jupyter K8s Iap Secret Name
-          invisible: true
-          section: jupyter_iap_auth
-        jupyter_k8s_ingress_name:
-          name: jupyter_k8s_ingress_name
-          title: Jupyter K8s Ingress Name
-          invisible: true
-          section: jupyter_iap_auth
-        jupyter_k8s_managed_cert_name:
-          name: jupyter_k8s_managed_cert_name
-          title: Jupyter K8s Managed Cert Name
-          invisible: true
-          section: jupyter_iap_auth
         jupyter_members_allowlist:
           name: jupyter_members_allowlist
           title: Jupyter Members Allowlist
@@ -310,35 +180,37 @@ spec:
           title: Rag Service Account
           invisible: true
           section: rag
+        ray_dashboard_add_auth:
+          name: ray_dashboard_add_auth
+          title: Enable IAP for Ray Dashboard Application
+          section: ray_iap_auth
+        ray_dashboard_domain:
+          name: ray_dashboard_domain
+          title: Ray Dashboard Domain
+          section: ray_iap_auth
+        ray_dashboard_members_allowlist:
+          name: ray_dashboard_members_allowlist
+          title: Ray Dashboard Members Allowlist
+          section: ray_iap_auth
         ray_service_account:
           name: ray_service_account
           title: Ray Service Account
           invisible: true
           section: rag
-        subnetwork_cidr:
-          name: subnetwork_cidr
-          title: Subnetwork Cidr
-          invisible: true
-          section: cluster_details
-        support_email:
-          name: support_email
-          title: Support Email
-          invisible: true
       sections:
         - name: acknowledge
           title: Before you begin
         - name: cluster_details
           title: GKE Cluster Configuration
-          tooltip: Select or Create GKE cluster
+          subtext: It is recommended to choose the <b>New Cluster</b> option
         - name: rag
           title: Other Configuration
         - name: rag_iap_auth
           title: Configure Authenticated Access for Frontend
-          subtext: Make sure the <a href="https://developers.google.com/workspace/guides/configure-oauth-consent#configure_oauth_consent"><i>OAuth Consent Screen</i></a> is configured for your project. Ensure <b>User type</b> is set to Internal.
+          subtext: Make sure the <a href="https://developers.google.com/workspace/guides/configure-oauth-consent#configure_oauth_consent"><i>OAuth Consent Screen</i></a> is configured for your project. Ensure <b>User type</b> is set to <i>Internal</i>. Note that by default, only users within your organization can be allowlisted. To add external users, change the <b>User type</b> to <i>External</i> after the application is deployed.
         - name: ray_iap_auth
           title: Configure Authenticated Access for Ray Dashboard
           subtext: Make sure the <a href="https://developers.google.com/workspace/guides/configure-oauth-consent#configure_oauth_consent"><i>OAuth Consent Screen</i></a> is configured for your project. Ensure <b>User type</b> is set to <i>Internal</i>. Note that by default, only users within your organization can be allowlisted. To add external users, change the <b>User type</b> to <i>External</i> after the application is deployed.
-
         - name: jupyter_iap_auth
           title: Configure Authenticated Access for JupyterHub
           subtext: Make sure the <a href="https://developers.google.com/workspace/guides/configure-oauth-consent#configure_oauth_consent"><i>OAuth Consent Screen</i></a> is configured for your project. Ensure <b>User type</b> is set to <i>Internal</i>. Note that by default, only users within your organization can be allowlisted. To add external users, change the <b>User type</b> to <i>External</i> after the application is deployed.
@@ -369,11 +241,6 @@ spec:
           openInNewTab: true
           showInNotification: true
           label: Go to Frontend Application
-        ray_dashboard_ip_address: {}
-        ray_dashboard_uri:
-          openInNewTab: true
-          showInNotification: true
-          label: Go to RAY Dashboard Application
         gcp_network: {}
         jupyterhub_ip_address: {}
         jupyterhub_password: {}
@@ -383,3 +250,6 @@ spec:
           label: Go to JupyterHub Application
         jupyterhub_user: {}
         kubernetes_namespace: {}
+        ray_dashboard_ip_address: {}
+        ray_dashboard_uri: {}
+