Update cert and missed Gateway API migration

Author: mattfsourcecode

README.md

@@ -44,4 +44,9 @@ This project showcases an educational and experimental setup, offering a startin
 - **Hybrid Cloud Deployments**: Adapt the setup for hybrid or multi-cloud Kubernetes deployments.
 - **Natural Language Processing (NLP)**: Implement AI-powered features such as text summarization, sentiment analysis, or chatbot functionality for applications requiring language understanding.
 - **Image and Video Processing**: Use AI models to enable facial recognition, object detection, image classification, or video analytics for multimedia applications.
-- **Image and Video Processing**: Use AI models to enable facial recognition, object detection, image classification, or video analytics for multimedia applications.
+- **Real-Time Data Stream Processing**: Integrate AI models to process and analyze high-velocity data streams (e.g., IoT sensor data, live event tracking, or financial market feeds) for real-time insights and predictions.
+- **AI-Powered Infrastructure Management**: Automate cluster health monitoring and resource allocation using predictive analytics to identify performance bottlenecks and self-heal infrastructure issues before they escalate.
+- **Scientific Simulations and Modeling**: Use AI to accelerate complex scientific simulations, such as climate modeling, molecular dynamics, or astrophysical computations, leveraging Kubernetes' scalable GPU resources.
+- **Context-Aware API Gateways**: Use AI models on Kubernetes endpoints to dynamically analyze incoming API requests and provide context-aware routing, such as adjusting traffic flow based on user behavior, request intent, or predicted resource demands. This can enhance scalability and improve user experience by intelligently prioritizing requests.
+- **Personalized Response Generation**: Deploy AI models on endpoints to deliver tailored responses to users, such as real-time content recommendations, adaptive UI/UX experiences, or personalized chatbot interactions. By integrating AI with Kubernetes, these models can scale based on traffic while ensuring low-latency, user-specific outputs for high-demand applications.
+- **Predictive Autoscaling for Endpoint Workloads**: Use AI models deployed on Kubernetes endpoints to predict traffic patterns and proactively scale resources. By analyzing historical and real-time data, the AI can optimize pod scaling to handle peak loads efficiently, reducing latency and preventing over-provisioning while ensuring seamless endpoint performance.

helm-chart/localai-values.yaml

@@ -8,27 +8,14 @@ service:
 
 # Gateway API disabled (using ingress instead)
 gateway:
-  enabled: false
-
-# Ingress configuration
-ingress:
   enabled: true
-  className: nginx
-  annotations:
-    cert-manager.io/cluster-issuer: selfsigned-issuer
-  hosts:
-    - host: example.local
-      paths:
-        - path: /
-          pathType: Prefix
-  tls:
-    - secretName: example-cert-tls
-      hosts:
-        - example.local
 
 # LocalAI configuration
 localai:
   enabled: true
+  env:
+    CONTEXT_SIZE: "0"
+    THREADS: "4"
   persistence:
     models:
       enabled: true

helm-chart/test-certificate.yaml

@@ -0,0 +1,17 @@
+apiVersion: cert-manager.io/v1
+kind: Certificate
+metadata:
+  name: test-cert
+  namespace: localai
+spec:
+  secretName: test-cert-tls
+  duration: 2160h # 90 days
+  renewBefore: 360h # 15 days
+  commonName: test.local
+  dnsNames:
+  - test.local
+  issuerRef:
+    name: selfsigned-issuer
+    kind: ClusterIssuer
+    group: cert-manager.io
+

helm-chart/values.yaml

@@ -1,5 +1,13 @@
 # LocalAI configuration
-fullnameOverride: "localai"
+
+certificate:
+  create: true
+  name: localai-tls
+  dnsNames:
+    - ai.example.com
+  issuerRef:
+    name: selfsigned-issuer
+    kind: ClusterIssuer
 
 autoscaling:
   enabled: false
@@ -83,10 +91,7 @@ gateway:
         mode: Terminate
         certificateRefs:
           - name: localai-tls
-            kind: Secret
-  hostnames:
-    - "ai.example.com"
-  rules:
+            rules:
     - matches:
         - path:
             type: PathPrefix