Varnish Cache

Deploy a Kubernetes-native HTTP caching layer with EDKA’s Varnish solution.

Kubernetes-Native Features

Automatic Service Discovery

• Service Endpoint Watching: Monitors Kubernetes services for real-time updates
• Dynamic Backend Updates: Services added/removed without restarts
• Transparent Proxy: Sits between Ingress and your applications
• Multi-Service Support: Handle multiple backend services simultaneously

Intelligent Cache Management

• BAN Support: Instant cache invalidation using regular expressions
• PURGE Support: Remove specific cached objects immediately
• Signaller Component: Broadcasts invalidation across all cache nodes
• Template-Based Control: Configure caching behavior via ConfigMap templates

Enterprise-Grade Reliability

• High Availability Mode: Multiple Varnish nodes with shared configuration
• StatefulSet Deployment: Persistent cache across pod restarts
• Load Balancing: Hash-based and round-robin routing strategies
• Health Probes: Kubernetes-native liveness and readiness checks

How It Works

1. Service-Based Architecture

Varnish sits between your Ingress controller and application services, acting as a transparent caching layer:

Ingress → Varnish Cache → Your Application

2. Dynamic Configuration via Templates

Configure caching behavior using Go templates in ConfigMaps - the controller automatically generates optimized VCL:

backend {{ .Name }} {
.host = "{{ .Host }}";
.port = "{{ .Port }}";
}

3. Automatic Endpoint Discovery

The controller watches Kubernetes service endpoints and automatically updates Varnish configuration when:

• Services are added or removed
• Pods scale up or down
• Backend endpoints change

Monitoring & Observability

Built-in Metrics

• Prometheus Integration: Export metrics in Prometheus format
• Cache Hit Ratio: Monitor cache effectiveness in real-time
• Request Analytics: Track requests, responses, and errors
• Backend Health: Monitor upstream service availability

Management Features

• kubectl Integration: Manage cache via standard Kubernetes tools
• Rolling Updates: Deploy changes without cache loss
• Backup & Restore: Persistent cache state management
• Multi-Region Support: Deploy across availability zones

Advanced Features

Cache Invalidation Strategies

• Cross-Node Signalling: Automatically propagate PURGE/BAN across all nodes
• Regex-Based BAN: Invalidate multiple URLs with pattern matching
• Direct PURGE: Remove specific cached objects instantly
• CI/CD Integration: Trigger cache invalidation from deployment pipelines

Security & Compliance

• Request Filtering: Block malicious requests before they reach backends
• Rate Limiting: Protect services from abuse and DDoS
• IP Whitelisting: Restrict access to specific IP ranges
• GDPR Compliance: Cookie handling and privacy headers

Developer-Friendly

• Template-Based Configuration: Use Go templates for flexible VCL generation
• Environment Variables: Support for variable interpolation in templates
• Custom VCL: Override default behavior with custom caching logic
• kubectl Compatible: Manage cache using standard Kubernetes tools

Use Cases

Microservices Architecture

• Cache responses between microservices
• Reduce inter-service latency
• Implement circuit breaker patterns
• Service mesh integration

Content Delivery

• Static asset acceleration
• Dynamic content caching with ESI
• Geo-distributed caching
• Mobile-optimized delivery

API Gateway

• Response caching for REST/GraphQL APIs
• Request coalescing for identical queries
• Backend protection during traffic spikes
• API versioning support

Performance Metrics

Based on real-world deployments:

• 99.9% Cache Availability: Enterprise-grade reliability
• <1ms Response Time: For cached content
• 90% Traffic Reduction: To backend services
• 10x Cost Savings: Reduced infrastructure requirements

Deploy Varnish