CloudHub vs Runtime Fabric: Complete Comparison Guide

Choosing between CloudHub and Runtime Fabric impacts your integration architecture significantly. This guide provides detailed comparison to help you make the right deployment decision.

Architecture Comparison

CloudHub Architecture

┌─────────────────────────────────────────────────────────────────┐
│                    ANYPOINT PLATFORM (SaaS)                     │
├─────────────────────────────────────────────────────────────────┤
│  ┌─────────────────────────────────────────────────────────┐   │
│  │                    CLOUDHUB                              │   │
│  │  ┌─────────┐  ┌─────────┐  ┌─────────┐  ┌─────────┐    │   │
│  │  │ Worker  │  │ Worker  │  │ Worker  │  │ Worker  │    │   │
│  │  │  0.1v   │  │  0.2v   │  │  1v     │  │  2v     │    │   │
│  │  └────┬────┘  └────┬────┘  └────┬────┘  └────┬────┘    │   │
│  │       │           │           │           │            │   │
│  │  ┌────▼───────────▼───────────▼───────────▼────┐      │   │
│  │  │              Shared Load Balancer            │      │   │
│  │  │        (*.cloudhub.io / *.us-e2.cloudhub.io) │      │   │
│  │  └─────────────────────────────────────────────┘      │   │
│  │                                                        │   │
│  │  Features:                                             │   │
│  │  ✓ Managed infrastructure                              │   │
│  │  ✓ Auto-scaling (CloudHub 2.0)                        │   │
│  │  ✓ Built-in monitoring                                 │   │
│  │  ✓ Persistent object store                             │   │
│  │  ✓ Shared/Dedicated load balancers                     │   │
│  └─────────────────────────────────────────────────────────┘   │
└─────────────────────────────────────────────────────────────────┘

Runtime Fabric Architecture

┌─────────────────────────────────────────────────────────────────┐
│                YOUR INFRASTRUCTURE (On-Prem/Cloud)              │
├─────────────────────────────────────────────────────────────────┤
│  ┌─────────────────────────────────────────────────────────┐   │
│  │              RUNTIME FABRIC (Kubernetes)                 │   │
│  │                                                          │   │
│  │  Controller Nodes (3)         Worker Nodes (N)          │   │
│  │  ┌─────────┐                 ┌─────────────────┐        │   │
│  │  │ RTF     │                 │ ┌─────┐ ┌─────┐ │        │   │
│  │  │ Agent   │◄───────────────►│ │Pod 1│ │Pod 2│ │        │   │
│  │  └─────────┘                 │ └─────┘ └─────┘ │        │   │
│  │  ┌─────────┐                 │ ┌─────┐ ┌─────┐ │        │   │
│  │  │ RTF     │                 │ │Pod 3│ │Pod 4│ │        │   │
│  │  │Deployer │                 │ └─────┘ └─────┘ │        │   │
│  │  └─────────┘                 └─────────────────┘        │   │
│  │                                                          │   │
│  │  ┌──────────────────────────────────────────────────┐   │   │
│  │  │            YOUR Load Balancer / Ingress           │   │   │
│  │  └──────────────────────────────────────────────────┘   │   │
│  │                                                          │   │
│  │  Features:                                               │   │
│  │  ✓ Full infrastructure control                          │   │
│  │  ✓ Custom networking                                    │   │
│  │  ✓ Data residency compliance                            │   │
│  │  ✓ Kubernetes-native                                    │   │
│  │  ✓ Air-gapped deployments possible                      │   │
│  └─────────────────────────────────────────────────────────┘   │
└─────────────────────────────────────────────────────────────────┘
           │
           ▼
┌─────────────────────────────────────────────────────────────────┐
│           ANYPOINT PLATFORM (Control Plane Only)                │
│           - Deployment management                               │
│           - Monitoring & analytics                              │
│           - API management                                      │
└─────────────────────────────────────────────────────────────────┘

Feature Comparison Matrix

| Feature | CloudHub | CloudHub 2.0 | Runtime Fabric | |---------|----------|--------------|----------------| | Deployment Model | iPaaS (SaaS) | iPaaS (SaaS) | Customer-managed K8s | | Infrastructure | MuleSoft managed | MuleSoft managed | Customer managed | | Scaling | Manual (workers) | Automatic HPA | Automatic HPA | | Min Replicas | 1 | 1 | 1 | | Max Replicas | 8 | 10 | Unlimited* | | Worker Sizes | 0.1v - 4v | Fractional | Custom CPU/Memory | | High Availability | Multi-worker | Multi-replica | Multi-replica + zones | | Load Balancer | Shared/Dedicated | Shared/Dedicated | Customer provided | | Persistent Storage | Object Store | Object Store | Customer provided | | Data Residency | Region-based | Region-based | Full control | | Network Isolation | VPC | VPC | Full control | | Custom DNS | Via DLB | Via DLB | Full control | | Air-gapped | No | No | Yes | | Monitoring | Built-in | Built-in | Built-in + custom |

Pricing Comparison

CloudHub Pricing Model

cloudhub_pricing:
  # vCore-based pricing
  worker_sizes:
    "0.1 vCore":
      cpu: "0.1"
      memory: "500MB"
      typical_use: "Development, testing"
 
    "0.2 vCore":
      cpu: "0.2"
      memory: "1GB"
      typical_use: "Light workloads"
 
    "1 vCore":
      cpu: "1"
      memory: "1.5GB"
      typical_use: "Standard production"
 
    "2 vCore":
      cpu: "2"
      memory: "3.5GB"
      typical_use: "High-throughput APIs"
 
    "4 vCore":
      cpu: "4"
      memory: "7.5GB"
      typical_use: "Heavy batch processing"
 
  # Dedicated Load Balancer (optional)
  dedicated_lb:
    base_cost: "Additional"
    features:
      - "Custom domain mapping"
      - "SSL certificate management"
      - "WAF integration"
      - "Custom routing rules"
 
  cost_optimization_tips:
    - "Use 0.1v workers for dev/test"
    - "Right-size based on actual usage"
    - "Share DLBs across apps when possible"
    - "Consider CloudHub 2.0 for auto-scaling"

Runtime Fabric Pricing Model

rtf_pricing:
  # Based on capacity (cores) allocated to RTF
  license_model: "Core-based capacity"
 
  infrastructure_costs:
    controller_nodes:
      count: 3  # Minimum for HA
      specification: "2 CPU, 8GB RAM each"
      your_cost: "Based on cloud/hardware provider"
 
    worker_nodes:
      count: "Variable based on workload"
      specification: "Custom based on needs"
      your_cost: "Based on cloud/hardware provider"
 
  additional_costs:
    - "Load balancer infrastructure"
    - "Persistent storage"
    - "Monitoring stack (if custom)"
    - "Network egress"
    - "Kubernetes management overhead"
 
  cost_optimization_tips:
    - "Use spot/preemptible instances for non-critical"
    - "Right-size nodes based on actual usage"
    - "Leverage auto-scaling to reduce idle capacity"
    - "Share RTF clusters across environments"
 
  total_cost_comparison:
    note: "RTF typically cost-effective at scale (50+ apps)"
    break_even: "Varies by cloud provider and workload"

Performance Comparison

Benchmark Configuration

// Performance test scenario
public class CloudHubVsRTFBenchmark {
 
    // Test configuration
    static final int CONCURRENT_USERS = 100;
    static final int REQUESTS_PER_USER = 1000;
    static final int PAYLOAD_SIZE_KB = 10;
 
    // CloudHub 2 vCore vs RTF 2 CPU
    @Test
    public void benchmarkThroughput() {
        // Results vary based on:
        // - Application complexity
        // - External dependencies
        // - Network latency
        // - Worker/pod placement
    }
}

Typical Performance Characteristics

performance_comparison:
  latency:
    cloudhub:
      p50: "50-100ms"
      p95: "150-300ms"
      p99: "300-500ms"
      notes:
        - "Shared infrastructure impact"
        - "Multi-tenant network"
        - "Geographic distance to region"
 
    runtime_fabric:
      p50: "20-50ms"
      p95: "50-150ms"
      p99: "100-250ms"
      notes:
        - "Dedicated resources"
        - "Local network"
        - "Customizable placement"
 
  throughput:
    cloudhub_2v:
      max_tps: "500-1000"  # Typical
      factors:
        - "Shared LB capacity"
        - "Worker isolation"
 
    rtf_2cpu:
      max_tps: "1000-2000"  # Typical
      factors:
        - "Dedicated resources"
        - "Custom LB configuration"
        - "Network proximity"
 
  scalability:
    cloudhub:
      horizontal: "Up to 8 workers"
      vertical: "Up to 4 vCore"
      scale_time: "2-5 minutes"
 
    runtime_fabric:
      horizontal: "Kubernetes limits (1000s)"
      vertical: "Node limits"
      scale_time: "30 seconds - 2 minutes"

Security Comparison

security_comparison:
  network_security:
    cloudhub:
      - "Anypoint VPC"
      - "VPN connectivity"
      - "VPC peering"
      - "Private Space (CloudHub 2.0)"
      limitations:
        - "Shared infrastructure"
        - "Limited custom firewall rules"
 
    runtime_fabric:
      - "Full network control"
      - "Custom security groups"
      - "Network policies"
      - "Service mesh integration"
      advantages:
        - "Zero-trust network capable"
        - "Custom ingress/egress rules"
        - "Air-gapped deployment"
 
  data_security:
    cloudhub:
      - "Encryption at rest (AES-256)"
      - "Encryption in transit (TLS 1.2+)"
      - "Secure properties"
      limitations:
        - "Keys managed by MuleSoft"
        - "Limited BYOK options"
 
    runtime_fabric:
      - "Full encryption control"
      - "BYOK (Bring Your Own Key)"
      - "Custom key management"
      - "HSM integration"
      advantages:
        - "Complete key lifecycle control"
        - "Custom encryption policies"
 
  compliance:
    cloudhub:
      certifications:
        - "SOC 2 Type II"
        - "ISO 27001"
        - "GDPR compliant regions"
        - "PCI DSS (shared responsibility)"
      limitations:
        - "Data residency per region"
        - "Limited audit controls"
 
    runtime_fabric:
      certifications:
        - "Inherits your certifications"
        - "Full audit trail"
        - "Custom compliance controls"
      advantages:
        - "Any data residency"
        - "Full audit capabilities"
        - "Custom compliance frameworks"

Decision Framework

Choose CloudHub When

cloudhub_ideal_for:
  scenarios:
    - "Quick time-to-market needed"
    - "Limited DevOps/infrastructure team"
    - "Smaller number of integrations (<50)"
    - "Standard compliance requirements"
    - "Predictable, moderate workloads"
    - "SaaS-first strategy"
 
  organization_profile:
    size: "SMB to Mid-Enterprise"
    it_maturity: "Low to Medium"
    cloud_strategy: "SaaS-preferred"
    compliance: "Standard (SOC2, ISO)"
 
  technical_requirements:
    data_residency: "Standard regions acceptable"
    network: "Internet-accessible or VPN"
    latency: "100-300ms acceptable"
    throughput: "<1000 TPS per app"
    scaling: "Predictable patterns"

Choose Runtime Fabric When

rtf_ideal_for:
  scenarios:
    - "Strict data residency requirements"
    - "Air-gapped or highly secure environments"
    - "Existing Kubernetes investment"
    - "Large-scale deployments (100+ apps)"
    - "Custom infrastructure requirements"
    - "High-performance, low-latency needs"
 
  organization_profile:
    size: "Enterprise"
    it_maturity: "High"
    cloud_strategy: "Multi-cloud or hybrid"
    compliance: "Strict (FedRAMP, HIPAA, custom)"
 
  technical_requirements:
    data_residency: "Specific locations required"
    network: "Private network, zero-trust"
    latency: "<50ms required"
    throughput: ">1000 TPS per app"
    scaling: "Unpredictable, burst patterns"
 
  prerequisites:
    - "Kubernetes expertise"
    - "Infrastructure automation"
    - "24/7 operations capability"
    - "Monitoring/logging stack"

Hybrid Approach

hybrid_deployment:
  description: "Use both based on workload characteristics"
 
  cloudhub_workloads:
    - "Customer-facing experience APIs"
    - "Partner integrations"
    - "Non-sensitive data processing"
    - "Variable/unpredictable workloads"
 
  rtf_workloads:
    - "Internal system APIs"
    - "Sensitive data processing"
    - "High-throughput batch jobs"
    - "Latency-critical integrations"
 
  architecture_example:
    experience_layer: "CloudHub (internet-facing)"
    process_layer: "Mixed (based on sensitivity)"
    system_layer: "RTF (internal, secure)"
 
  considerations:
    - "Network connectivity between environments"
    - "Consistent monitoring approach"
    - "Unified API management"
    - "CI/CD pipeline complexity"

Migration Considerations

CloudHub to RTF Migration

#!/bin/bash
# Migration checklist script
 
echo "=== CloudHub to RTF Migration Checklist ==="
 
# Application inventory
echo "1. Application Inventory"
echo "   - List all CloudHub applications"
echo "   - Document vCore allocations"
echo "   - Identify dependencies"
 
# Infrastructure preparation
echo "2. Infrastructure Preparation"
echo "   - Provision Kubernetes cluster"
echo "   - Install RTF"
echo "   - Configure networking"
echo "   - Set up persistent storage"
 
# Configuration migration
echo "3. Configuration Migration"
echo "   - Export secure properties"
echo "   - Migrate to Kubernetes secrets"
echo "   - Update endpoint configurations"
echo "   - Configure ingress/load balancer"
 
# Testing
echo "4. Testing Phase"
echo "   - Deploy to RTF staging"
echo "   - Validate functionality"
echo "   - Performance testing"
echo "   - Security scanning"
 
# Cutover
echo "5. Cutover"
echo "   - DNS/routing update"
echo "   - Traffic migration"
echo "   - Monitoring validation"
echo "   - Rollback plan ready"

Conclusion

CloudHub offers simplicity and managed infrastructure ideal for organizations prioritizing speed and minimal operations overhead. Runtime Fabric provides control and customization for enterprises with strict compliance requirements or existing Kubernetes investments. Many organizations benefit from a hybrid approach, placing workloads where they fit best. Evaluate based on your specific requirements around compliance, performance, scale, and operational capabilities.