Behind the Scenes of Google Cloud Networking
Ever wondered how cloud networking works behind the scenes?
In cloud environments, When we create virtual networks, firewalls, route tables, etc.
It happens in a matter of seconds to minutes.
It's not any magic; there is hardware sitting at some place in the data center. All those complex configurations are abstracted by virtualization technologies, Sofware-defined networking, and cloud providers proprietary software.
Let's take the example of Google cloud.
When it comes to Google Cloud, its cloud network virtualization stack using Andromeda SDN.
It is the orchestration point for provisioning, configuring, and managing virtual networks and in-network packet processing.
All the cloud platform networking services with high performance, availability, isolation, and security are delivered using Andromeda.
For example, when you create firewalls, routing, and forwarding rules on google cloud, it uses the backend Andromeda APIs and infrastructure.
Google first announced Andromeda in 2014.
In 2019, it released Andromeda 2.2 for high-throughput VMs.
Few more information on Google Cloud Data Cernter.
Google cloud uses Clos topology for its network.
Centralized software control stack to manage all switches within the data center.
Google doesn't rely on standard internet protocols. Instead, they build their custom software and hardware tailored to the data center.
Sources & References
[1]. Google Cloud Platform's latest networking stack
[2]. How Andromeda 2.2 enables high-throughput VMs
[3]. A look inside Google’s Data Center Networks
[4]. A Decade of Clos Topologies and Centralized Control in Google’s Datacenter Network