feat: add prometheus docker service for node performance metrics analytics

.env.example

@@ -39,6 +39,12 @@ COMPOSE_PROFILES=chat_completions_vllm
 # Tracing level
 TRACE_LEVEL=info
 
+# Prometheus Configuration
+PROMETHEUS_PORT=9090
+
+# Grafana Configuration
+GRAFANA_PORT=30001
+
 # ----------------------------------------------------------------------------------
 # chat completions server
 CHAT_COMPLETIONS_SERVER_PORT=50000

atoma-service/docs/openapi.yml

@@ -12,7 +12,7 @@ paths:
     get:
       tags:
       - health
-      summary: Handles the health check endpoint.
+      summary: Health
       description: |-
         This function is used to verify that the server is running and responsive.
         It's typically used by load balancers or monitoring systems to check the
@@ -41,7 +41,7 @@ paths:
     get:
       tags:
       - metrics
-      summary: Handles the metrics endpoint.
+      summary: Metrics
       description: |-
         This function is used to return the metrics for the service.
 
@@ -56,7 +56,7 @@ paths:
     post:
       tags:
       - chat
-      summary: Handles chat completion requests by forwarding them to the inference service and managing token usage.
+      summary: Create chat completion
       description: |-
         This handler performs several key operations:
         1. Forwards the chat completion request to the inference service
@@ -102,8 +102,10 @@ paths:
     post:
       tags:
       - embeddings
-      summary: Handles embedding requests by proxying them to the embeddings service.
-      description: |-
+      summary: Create embeddings
+      description: |2-
+
+
         This handler simply forwards the request to the embeddings service and returns the response.
 
         # Arguments
@@ -138,8 +140,10 @@ paths:
     post:
       tags:
       - images
-      summary: Handles image generation requests by proxying them to the image generations service.
-      description: |-
+      summary: Create image generation
+      description: |2-
+
+
         This handler simply forwards the request to the image generations service and returns the response.
 
         # Arguments

docker-compose.yaml

@@ -61,6 +61,30 @@ services:
     networks:
       - atoma-network
 
+  prometheus:
+    image: prom/prometheus:latest
+    restart: always
+    ports:
+      - "${PROMETHEUS_PORT:-9090}:9090"
+    volumes:
+      - ./prometheus.yml:/etc/prometheus/prometheus.yml
+    env_file: .env
+    networks:
+      - atoma-network
+
+  grafana:
+    image: grafana/grafana:latest
+    restart: always
+    ports:
+      - "${GRAFANA_PORT:-30001}:30000"
+    depends_on:
+      - prometheus
+    volumes:
+      - grafana_data:/var/lib/grafana
+    env_file: .env
+    networks:
+      - atoma-network
+
   atoma-node:
     <<: *atoma-node
     build:
@@ -166,3 +190,4 @@ networks:
 
 volumes:
   postgres-data:
+  grafana_data:

prometheus.yml

@@ -0,0 +1,9 @@
+global:
+  scrape_interval: 15s
+  evaluation_interval: 15s
+
+scrape_configs:
+  - job_name: 'atoma-node'
+    metrics_path: '/metrics'
+    static_configs:
+      - targets: ['atoma-node:3000']