hybrid-cloud-networking

---
name: hybrid-cloud-networking
description: "Configure secure, high-performance connectivity between on-premises and cloud environments using VPN, Direct Connect, and ExpressRoute."
risk: safe
source: community
date_added: "2026-02-27"
---

# Hybrid Cloud Networking

Configure secure, high-performance connectivity between on-premises and cloud environments using VPN, Direct Connect, and ExpressRoute.

## Do not use this skill when

- The task is unrelated to hybrid cloud networking
- You need a different domain or tool outside this scope

## Instructions

- Clarify goals, constraints, and required inputs.
- Apply relevant best practices and validate outcomes.
- Provide actionable steps and verification.
- If detailed examples are required, open `resources/implementation-playbook.md`.

## Purpose

Establish secure, reliable network connectivity between on-premises data centers and cloud providers (AWS, Azure, GCP).

## Use this skill when

- Connect on-premises to cloud
- Extend datacenter to cloud
- Implement hybrid active-active setups
- Meet compliance requirements
- Migrate to cloud gradually

## Connection Options

### AWS Connectivity

#### 1. Site-to-Site VPN
- IPSec VPN over internet
- Up to 1.25 Gbps per tunnel
- Cost-effective for moderate bandwidth
- Higher latency, internet-dependent

```hcl
resource "aws_vpn_gateway" "main" {
  vpc_id = aws_vpc.main.id
  tags = {
    Name = "main-vpn-gateway"
  }
}

resource "aws_customer_gateway" "main" {
  bgp_asn    = 65000
  ip_address = "203.0.113.1"
  type       = "ipsec.1"
}

resource "aws_vpn_connection" "main" {
  vpn_gateway_id      = aws_vpn_gateway.main.id
  customer_gateway_id = aws_customer_gateway.main.id
  type                = "ipsec.1"
  static_routes_only  = false
}
```

#### 2. AWS Direct Connect
- Dedicated network connection
- 1 Gbps to 100 Gbps
- Lower latency, consistent bandwidth
- More expensive, setup time required

**Reference:** See `references/direct-connect.md`

### Azure Connectivity

#### 1. Site-to-Site VPN
```hcl
resource "azurerm_virtual_network_gateway" "vpn" {
  name                = "vpn-gateway"
  location            = azurerm_resource_group.main.location
  resource_group_name = azurerm_resource_group.main.name

  type     = "Vpn"
  vpn_type = "RouteBased"
  sku      = "VpnGw1"

  ip_configuration {
    name                          = "vnetGatewayConfig"
    public_ip_address_id          = azurerm_public_ip.vpn.id
    private_ip_address_allocation = "Dynamic"
    subnet_id                     = azurerm_subnet.gateway.id
  }
}
```

#### 2. Azure ExpressRoute
- Private connection via connectivity provider
- Up to 100 Gbps
- Low latency, high reliability
- Premium for global connectivity

### GCP Connectivity

#### 1. Cloud VPN
- IPSec VPN (Classic or HA VPN)
- HA VPN: 99.99% SLA
- Up to 3 Gbps per tunnel

#### 2. Cloud Interconnect
- Dedicated (10 Gbps, 100 Gbps)
- Partner (50 Mbps to 50 Gbps)
- Lower latency than VPN

## Hybrid Network Patterns

### Pattern 1: Hub-and-Spoke
```
On-Premises Datacenter
         ↓
    VPN/Direct Connect
         ↓
    Transit Gateway (AWS) / vWAN (Azure)
         ↓
    ├─ Production VPC/VNet
    ├─ Staging VPC/VNet
    └─ Development VPC/VNet
```

### Pattern 2: Multi-Region Hybrid
```
On-Premises
    ├─ Direct Connect → us-east-1
    └─ Direct Connect → us-west-2
            ↓
        Cross-Region Peering
```

### Pattern 3: Multi-Cloud Hybrid
```
On-Premises Datacenter
    ├─ Direct Connect → AWS
    ├─ ExpressRoute → Azure
    └─ Interconnect → GCP
```

## Routing Configuration

### BGP Configuration
```
On-Premises Router:
- AS Number: 65000
- Advertise: 10.0.0.0/8

Cloud Router:
- AS Number: 64512 (AWS), 65515 (Azure)
- Advertise: Cloud VPC/VNet CIDRs
```

### Route Propagation
- Enable route propagation on route tables
- Use BGP for dynamic routing
- Implement route filtering
- Monitor route advertisements

## Security Best Practices

1. **Use private connectivity** (Direct Connect/ExpressRoute)
2. **Implement encryption** for VPN tunnels
3. **Use VPC endpoints** to avoid internet routing
4. **Configure network ACLs** and security groups
5. **Enable VPC Flow Logs** for monitoring
6. **Implement DDoS protection**
7. **Use PrivateLink/Private Endpoints**
8. **Monitor connections** with CloudWatch/Monitor
9. **Implement redundancy** (dual tunnels)
10. **Regular security audits**

## High Availability

### Dual VPN Tunnels
```hcl
resource "aws_vpn_connection" "primary" {
  vpn_gateway_id      = aws_vpn_gateway.main.id
  customer_gateway_id = aws_customer_gateway.primary.id
  type                = "ipsec.1"
}

resource "aws_vpn_connection" "secondary" {
  vpn_gateway_id      = aws_vpn_gateway.main.id
  customer_gateway_id = aws_customer_gateway.secondary.id
  type                = "ipsec.1"
}
```

### Active-Active Configuration
- Multiple connections from different locations
- BGP for automatic failover
- Equal-cost multi-path (ECMP) routing
- Monitor health of all connections

## Monitoring and Troubleshooting

### Key Metrics
- Tunnel status (up/down)
- Bytes in/out
- Packet loss
- Latency
- BGP session status

### Troubleshooting
```bash
# AWS VPN
aws ec2 describe-vpn-connections
aws ec2 get-vpn-connection-telemetry

# Azure VPN
az network vpn-connection show
az network vpn-connection show-device-config-script
```

## Cost Optimization

1. **Right-size connections** based on traffic
2. **Use VPN for low-bandwidth** workloads
3. **Consolidate traffic** through fewer connections
4. **Minimize data transfer** costs
5. **Use Direct Connect** for high bandwidth
6. **Implement caching** to reduce traffic

## Reference Files

- `references/vpn-setup.md` - VPN configuration guide
- `references/direct-connect.md` - Direct Connect setup

## Related Skills

- `multi-cloud-architecture` - For architecture decisions
- `terraform-module-library` - For IaC implementation

## Limitations
- Use this skill only when the task clearly matches the scope described above.
- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

使用 Skill 时，您的对话内容将被发送至 AI 模型进行处理。我们会严格保护您的隐私数据，不会将您的对话内容用于模型训练或分享给第三方。 以下为此 Skill 的数据处理说明。