Intrigued by the art of cloud architecture? Discover how to design, develop, and manage robust, secure, scalable, and dynamic solutions on Google Cloud as you prepare for the Professional Cloud Architect exam!
Prepare and test your skills
Prepare and test your skills
To connect different projects privately, you can use VPC Network Peering. This allows two VPC networks to communicate using internal IP addresses, keeping traffic off the public internet. For centralized control, Shared VPC lets a host project manage the network while allowing other service projects to use it. This separation helps maintain security by keeping network administration separate from application teams.
Private Service Connect provides a way to privately access managed services and APIs. It uses a consumer-producer model where a service is exposed through a single endpoint in the consumer's VPC. This eliminates the need for public IPs or complex peering. Key components are Private Endpoints, which use internal IPs, and Service Attachments, which allow producers to publish their services.
For large, complex environments, Network Connectivity Center acts as a central hub. It supports hub-and-spoke topologies to connect VPCs, on-premises data centers, and other clouds. Using this hub simplifies how routes are shared and managed across a global network, which is essential for scaling multicloud architectures.
A system architecture diagram showing a central Network Connectivity Center hub in Google Cloud connected to various spokes, including on-premises data centers via Cloud Interconnect, other cloud providers via Cross-Cloud Interconnect, and internal VPC networks.
To connect Google Cloud to on-premises or other clouds, you have two main options. Cloud Interconnect provides a high-bandwidth, physical link, while Cloud VPN creates encrypted tunnels over the internet. Dedicated Interconnect is a direct connection to Google, and Partner Interconnect goes through a service provider. HA VPN provides high-availability tunnels for critical connections.
Security in a shared environment is enforced by VPC Service Controls. They create a security perimeter that prevents data from being moved outside authorized projects. Cloud Service Mesh implements a zero-trust architecture by using identities and encryption for every service-to-service request. Together, they ensure only authorized workloads can communicate, even across different organizations.
Cloud Router manages dynamic routing within and between networks. It uses the Border Gateway Protocol (BGP) to automatically exchange route information with external networks, like on-premises data centers or other clouds. This ensures traffic always finds the best available path without manual updates, which is crucial for maintaining stable connections in complex architectures.
High Availability (HA) VPN is used to securely connect Google Cloud to other environments. To achieve a 99.99% SLA, you configure two VPN tunnels from an HA VPN gateway. These tunnels work with a Cloud Router to handle traffic failover. When connecting, you must define an External VPN Gateway resource to represent the peer side and choose the correct redundancy type (like SINGLE_IP_INTERNALLY_REDUNDANT or TWO_IPS_REDUNDANCY) based on the peer's capabilities.
Cloud NAT allows resources without external IPs to access the internet while blocking incoming connections. You can choose automatic or manual IP allocation. Manual allocation is needed when you must provide specific, fixed IP addresses to external partners. Cloud NAT can also integrate with Managed Proxy Load Balancers using a proxy-only subnet to manage traffic efficiently.
In complex multicloud setups, a Transit VPC often acts as a central hub for all traffic. This allows you to place Network Virtual Appliances (NVAs) for security in one central location instead of every network. A single-NIC NVA deployment is recommended because it simplifies scaling and allows for flexible traffic steering using policy-based routes. This centralized approach ensures consistent security policies and supports high availability through an internal load balancer.
Selecting the right connection depends on your needs for bandwidth, cost, and reliability. Cloud VPN is a cost-effective option that creates encrypted tunnels over the public internet, suitable for lower traffic volumes. For higher performance, Cloud Interconnect provides a private, high-speed link. It comes in two forms: Dedicated Interconnect is a direct physical connection, and Partner Interconnect connects through a supported service provider.
For direct connections between Google Cloud and other clouds like AWS or Azure, Cross-Cloud Interconnect is essential. It establishes direct, managed connections without extra hardware. This reduces latency, improves security, and provides high-bandwidth communication, helping avoid vendor lock-in.
Cloud Router handles dynamic routing using BGP to automatically update network paths. For managing many connections, the Network Connectivity Center acts as a central hub. This ensures smooth traffic flow between on-premises sites, Google Cloud, and other cloud environments.
While Interconnect is private, you can add HA VPN over Cloud Interconnect for extra IPsec encryption. Google provides a Service Level Agreement (SLA) of up to 99.99% for highly available configurations. Using these tools together ensures the multicloud environment is secure, scalable, and resilient against failures.
VPC Network Peering allows two VPC networks to communicate using internal IP addresses, keeping traffic off the public internet, while Shared VPC lets a host project manage the network and allows other service projects to use it for centralized control.
Cloud VPN is a cost-effective option that creates encrypted tunnels over the public internet, suitable for lower traffic volumes, while Cloud Interconnect provides a private, high-speed link for higher performance needs.
Network Connectivity Center acts as a central hub that supports hub-and-spoke topologies to connect VPCs, on-premises data centers, and other clouds, simplifying how routes are shared and managed across a global network.
Dedicated Interconnect is a direct physical connection to Google, and Partner Interconnect connects through a supported service provider.
An enterprise is deploying a multicloud architecture connecting its Google Cloud Virtual Private Cloud (VPC) network directly to workloads hosted in Amazon Web Services (AWS).
The networking team defines the following requirements and constraints:
Which hybrid and multicloud connectivity solution should the enterprise choose?
Provision Google Cloud Cross-Cloud Interconnect connections to AWS.
Configure Cloud HA VPN tunnels over the public internet to the remote cloud provider.
Establish Dedicated Interconnect by leasing space in a shared colocation facility.
Deploy Partner Interconnect through a multicloud-enabled network service provider.