Feat/major refactor

.gitignore

@@ -57,7 +57,7 @@ docker-compose.override.yml
 *.db
 *.sqlite
 *.sqlite3
-data/
+
 
 # Cache and temporary files
 .cache/
@@ -94,12 +94,13 @@ docs/.vitepress/dist/
 docs/.vitepress/cache/
 
 # Generated Swagger documentation
-internal/handlers/swagger/docs.go
-internal/handlers/swagger/swagger.json
-internal/handlers/swagger/swagger.yaml
+internal/api/handlers/swagger/docs.go
+internal/api/handlers/swagger/swagger.json
+internal/api/handlers/swagger/swagger.yaml
 
 # Embedded documentation dist
-internal/docs/dist/
+internal/api/docs/dist/
+internal/api/ui/dist/
 
 # Claude specific files
 .claude/

Dockerfile

@@ -43,7 +43,7 @@ COPY docs/ ./
 RUN npm run build
 
 # Go Build stage
-FROM golang:1.23-alpine AS builder
+FROM golang:1.24-alpine AS builder
 
 # Install build dependencies
 RUN apk add --no-cache git gcc musl-dev
@@ -61,14 +61,14 @@ RUN go mod download
 COPY . .
 
 # Copy built UI from ui-builder stage
-COPY --from=ui-builder /app/web/dist ./internal/ui/dist
+COPY --from=ui-builder /app/web/dist ./internal/api/ui/dist
 
 # Copy built docs from docs-builder stage
-COPY --from=docs-builder /app/docs/.vitepress/dist ./internal/docs/dist
+COPY --from=docs-builder /app/docs/.vitepress/dist ./internal/api/docs/dist
 
 # Generate Swagger documentation
 RUN go install github.com/swaggo/swag/cmd/swag@latest
-RUN swag init -g cmd/server/main.go -o internal/handlers/swagger
+RUN swag init -g cmd/server/main.go -o internal/api/handlers/swagger
 
 # Build the application with embedded UI
 RUN CGO_ENABLED=1 GOOS=linux go build -a -installsuffix cgo -o pllm cmd/server/main.go
@@ -88,13 +88,12 @@ WORKDIR /app
 
 # Copy binary from builder
 COPY --from=builder /app/pllm .
-COPY --from=builder /app/docs ./docs
 
 # Copy config file (if exists)
 COPY --chown=pllm:pllm config.yaml* ./
 
 # Copy pricing file
-COPY --from=builder --chown=pllm:pllm /app/internal/config/model_prices_and_context_window.json ./internal/config/
+COPY --from=builder --chown=pllm:pllm /app/internal/core/config/model_prices_and_context_window.json ./internal/core/config/
 
 # Change ownership
 RUN chown -R pllm:pllm /app
@@ -110,4 +109,4 @@ HEALTHCHECK --interval=30s --timeout=3s --start-period=5s --retries=3 \
     CMD wget --no-verbose --tries=1 --spider http://localhost:8080/health || exit 1
 
 # Run the application
-ENTRYPOINT ["./pllm"]
+ENTRYPOINT ["./pllm"]

Makefile

@@ -27,10 +27,10 @@ web-build: ## Build frontend assets
 	cd web && npm run build
 
 .PHONY: ui-build
-ui-build: web-build ## Copy built frontend to internal/ui/dist for embedding
-	@mkdir -p internal/ui/dist
-	@cp -r web/dist/* internal/ui/dist/
-	@echo "✅ Frontend copied to internal/ui/dist/"
+ui-build: web-build ## Copy built frontend to internal/api/ui/dist for embedding
+	@mkdir -p internal/api/ui/dist
+	@cp -r web/dist/* internal/api/ui/dist/
+	@echo "✅ Frontend copied to internal/api/ui/dist/"
 
 .PHONY: build-worker
 build-worker: ## Build the worker binary for background processing
@@ -336,24 +336,32 @@ redis-shell: ## Open Redis shell
 
 .PHONY: test
 test: swagger ## Run tests (generates swagger docs first)
-	mkdir -p internal/ui/dist
-	mkdir -p internal/docs/dist
-	touch internal/ui/dist/index.html
-	touch internal/docs/dist/index.html
+	mkdir -p internal/api/ui/dist
+	mkdir -p internal/api/docs/dist
+	touch internal/api/ui/dist/index.html
+	touch internal/api/docs/dist/index.html
 	go test -v ./...
 
 .PHONY: test-coverage
 test-coverage: swagger ## Run tests with coverage
-	mkdir -p internal/ui/dist
-	mkdir -p internal/docs/dist
-	touch internal/ui/dist/index.html
-	touch internal/docs/dist/index.html
+	mkdir -p internal/api/ui/dist
+	mkdir -p internal/api/docs/dist
+	touch internal/api/ui/dist/index.html
+	touch internal/api/docs/dist/index.html
 	go test -v -cover -coverprofile=coverage.txt ./...
 
 .PHONY: test-integration
 test-integration: ## Run integration tests
 	go test -v -tags=integration ./...
 
+.PHONY: test-failover
+test-failover: ## Run failover and performance integration tests
+	go test -v -timeout=60s ./internal/services/integration/ -run="Test"
+
+.PHONY: test-performance
+test-performance: ## Run performance benchmarks and validate banking requirements
+	go test -v -timeout=60s ./internal/services/integration/ -run="TestPerformanceBenchmarks"
+
 .PHONY: lint
 lint: ## Run linter
 	@which golangci-lint > /dev/null || go install github.com/golangci/golangci-lint/cmd/golangci-lint@latest
@@ -375,7 +383,7 @@ install-tools: ## Install development tools
 .PHONY: swagger
 swagger: ## Generate Swagger documentation
 	@which swag > /dev/null || go install github.com/swaggo/swag/cmd/swag@latest
-	swag init -g cmd/server/main.go -o internal/handlers/swagger
+	swag init -g cmd/server/main.go -o internal/api/handlers/swagger
 
 ##@ Documentation
 
@@ -386,16 +394,16 @@ docs-dev: ## Run VitePress documentation in development mode
 .PHONY: docs-build
 docs-build: ## Build VitePress documentation
 	cd docs && npm run build
-	mkdir -p internal/docs/dist
-	cp -r docs/.vitepress/dist/* internal/docs/dist/
+	mkdir -p internal/api/docs/dist
+	cp -r docs/.vitepress/dist/* internal/api/docs/dist/
 
 .PHONY: docs-preview
 docs-preview: ## Preview built documentation
 	cd docs && npm run preview
 
 .PHONY: clean
 clean: ## Clean build artifacts
-	rm -rf bin/ tmp/ coverage.* *.out internal/docs/dist
+	rm -rf bin/ tmp/ coverage.* *.out internal/api/docs/dist
 
 .PHONY: env-setup
 env-setup: ## Create .env file from example

cmd/pllm/commands/budget.go → cmd/cli/commands/budget.go

@@ -10,7 +10,7 @@ import (
 	"github.com/google/uuid"
 	"github.com/spf13/cobra"
 
-	"github.com/amerfu/pllm/internal/models"
+	"github.com/amerfu/pllm/internal/core/models"
 )
 
 // NewBudgetCommand creates a new budget management command

cmd/pllm/commands/key.go → cmd/cli/commands/key.go

@@ -9,7 +9,7 @@ import (
 	"github.com/google/uuid"
 	"github.com/spf13/cobra"
 
-	"github.com/amerfu/pllm/internal/models"
+	"github.com/amerfu/pllm/internal/core/models"
 )
 
 // NewKeyCommand creates a new key management command

cmd/pllm/commands/team.go → cmd/cli/commands/team.go

@@ -9,7 +9,7 @@ import (
 	"github.com/google/uuid"
 	"github.com/spf13/cobra"
 
-	"github.com/amerfu/pllm/internal/models"
+	"github.com/amerfu/pllm/internal/core/models"
 )
 
 // NewTeamCommand creates a new team management command

cmd/pllm/commands/user.go → cmd/cli/commands/user.go

@@ -9,7 +9,7 @@ import (
 	"github.com/google/uuid"
 	"github.com/spf13/cobra"
 
-	"github.com/amerfu/pllm/internal/models"
+	"github.com/amerfu/pllm/internal/core/models"
 )
 
 // NewUserCommand creates a new user management command

cmd/pllm/main.go → cmd/cli/main.go

@@ -9,8 +9,8 @@ import (
 	"gorm.io/driver/postgres"
 	"gorm.io/gorm"
 
-	"github.com/amerfu/pllm/cmd/pllm/commands"
-	"github.com/amerfu/pllm/internal/models"
+	"github.com/amerfu/pllm/cmd/cli/commands"
+	"github.com/amerfu/pllm/internal/core/models"
 )
 
 var (

cmd/server/main.go

@@ -9,20 +9,20 @@ import (
 	"syscall"
 	"time"
 
-	"github.com/amerfu/pllm/internal/config"
-	"github.com/amerfu/pllm/internal/database"
-	"github.com/amerfu/pllm/internal/logger"
-	"github.com/amerfu/pllm/internal/router"
-	"github.com/amerfu/pllm/internal/services/cache"
-	"github.com/amerfu/pllm/internal/services/models"
-	redisService "github.com/amerfu/pllm/internal/services/redis"
+	"github.com/amerfu/pllm/internal/core/config"
+	"github.com/amerfu/pllm/internal/core/database"
+	"github.com/amerfu/pllm/pkg/logger"
+	"github.com/amerfu/pllm/internal/api/router"
+	"github.com/amerfu/pllm/internal/services/data/cache"
+	"github.com/amerfu/pllm/internal/services/llm/models"
+	redisService "github.com/amerfu/pllm/internal/services/data/redis"
 	"github.com/amerfu/pllm/internal/services/worker"
 	"github.com/joho/godotenv"
 	"github.com/redis/go-redis/v9"
 	"go.uber.org/zap"
 	"gorm.io/gorm"
 
-	_ "github.com/amerfu/pllm/internal/handlers/swagger"
+	// _ "github.com/amerfu/pllm/internal/api/handlers/swagger" // TODO: Generate swagger docs
 )
 
 // @title pllm - Blazing Fast LLM Gateway
@@ -138,8 +138,38 @@ func main() {
 	// Services are now initialized in the router
 	// All authentication and management functionality is handled by the unified auth service
 
+	// Initialize Redis client early for model manager and worker
+	var redisClient *redis.Client
+	if appMode.RedisAvailable {
+		opt, err := redis.ParseURL(cfg.Redis.URL)
+		if err != nil {
+			log.Warn("Failed to parse Redis URL, continuing without Redis", zap.Error(err))
+			appMode.RedisAvailable = false
+		} else {
+			// Override with explicit password and DB if provided
+			if cfg.Redis.Password != "" {
+				opt.Password = cfg.Redis.Password
+			}
+			if cfg.Redis.DB != 0 {
+				opt.DB = cfg.Redis.DB
+			}
+
+			redisClient = redis.NewClient(opt)
+
+			// Test Redis connection
+			if err := redisClient.Ping(context.Background()).Err(); err != nil {
+				log.Warn("Redis not available, continuing without Redis features", zap.Error(err))
+				redisClient = nil
+				appMode.RedisAvailable = false
+			} else {
+				log.Info("Redis connected successfully")
+			}
+		}
+	}
+
 	// Initialize model manager (always needed)
-	modelManager := models.NewModelManager(log, cfg.Router)
+	// Pass Redis client for distributed latency tracking (nil if Redis not available)
+	modelManager := models.NewModelManager(log, cfg.Router, redisClient)
 	if err := modelManager.LoadModelInstances(cfg.ModelList); err != nil {
 		log.Fatal("Failed to load model instances", zap.Error(err))
 	}
@@ -166,27 +196,10 @@ func main() {
 	var workerCtx context.Context
 	var workerCancel context.CancelFunc
 
-	if !appMode.IsLiteMode && appMode.RedisAvailable && db != nil {
-		// Initialize Redis client for worker
-		opt, err := redis.ParseURL(cfg.Redis.URL)
-		if err != nil {
-			log.Fatal("Failed to parse Redis URL", zap.Error(err))
-		}
-
-		// Override with explicit password and DB if provided
-		if cfg.Redis.Password != "" {
-			opt.Password = cfg.Redis.Password
-		}
-		if cfg.Redis.DB != 0 {
-			opt.DB = cfg.Redis.DB
-		}
-
-		redisClient := redis.NewClient(opt)
-
-		// Test Redis connection for worker
-		if err := redisClient.Ping(context.Background()).Err(); err != nil {
-			log.Warn("Redis not available for background worker", zap.Error(err))
-		} else {
+	if !appMode.IsLiteMode && appMode.RedisAvailable && db != nil && redisClient != nil {
+		// Use the Redis client initialized earlier
+		// Redis connection already verified above
+		{
 			// Initialize Redis services for worker
 			usageQueue := redisService.NewUsageQueue(&redisService.UsageQueueConfig{
 				Client:     redisClient,

cmd/worker/main.go

@@ -14,8 +14,8 @@ import (
 	"gorm.io/driver/postgres"
 	"gorm.io/gorm"
 
-	"github.com/amerfu/pllm/internal/config"
-	redisService "github.com/amerfu/pllm/internal/services/redis"
+	"github.com/amerfu/pllm/internal/core/config"
+	redisService "github.com/amerfu/pllm/internal/services/data/redis"
 	"github.com/amerfu/pllm/internal/services/worker"
 )
 

docs/.vitepress/config.mjs

@@ -16,37 +16,39 @@ export default withMermaid(
       ],
       sidebar: [
         {
-          text: "Getting Started",
+          text: "Introduction",
           items: [
-            { text: "What is pllm?", link: "/" },
-            { text: "Installation & Setup", link: "/guide/getting-started" },
-            { text: "Quick Start Guide", link: "/guide/quickstart" },
+            { text: "What is PLLM?", link: "/" },
+            { text: "Quick Start", link: "/guide/quickstart" },
+            { text: "Installation", link: "/guide/getting-started" },
           ],
         },
         {
-          text: "Core Features",
+          text: "Configuration",
           items: [
-            { text: "System Architecture", link: "/guide/architecture" },
-            { text: "Multi-Provider Support", link: "/providers" },
+            { text: "Configuration Guide", link: "/config" },
+            { text: "Model Routing & Load Balancing", link: "/guide/routing" },
+            { text: "Provider Setup", link: "/providers" },
             { text: "Authentication", link: "/auth" },
-            { text: "Configuration", link: "/config" },
+          ],
+        },
+        {
+          text: "Architecture",
+          items: [
+            { text: "System Overview", link: "/guide/architecture" },
+            { text: "Resilience & Reliability", link: "/guide/resilience" },
           ],
         },
         {
           text: "API Reference",
           items: [
             { text: "OpenAI Compatible API", link: "/api" },
-            { text: "Chat Completions", link: "/api#chat-completions" },
-            { text: "Models", link: "/api#models" },
-            { text: "Health Checks", link: "/api#health-checks" },
           ],
         },
         {
           text: "Deployment",
           items: [
-            { text: "Docker Deployment", link: "/deployment" },
-            { text: "Kubernetes", link: "/deployment#kubernetes" },
-            { text: "Production Setup", link: "/deployment#production" },
+            { text: "Docker & Kubernetes", link: "/deployment" },
           ],
         },
       ],