There are very specific services that you are engaging with as you traverse from where you are to the provider's edge routers, through their edge switches, aggregation switches, access switches, virtual switches and, finally, to your new virtual Machine.
(Note: This is a generic discussion of the networking that goes into a public cloud infrastructure. Therefore, I am not referencing any one provider's network architecture.)
Because providers don't know where you are coming from, they need to be able to allow access from a variety of sources. The edge services for the public cloud provider would allow access from a wide range of users, residential and business applications, and services, including high-speeed transport and VPN services, next-generation broadband multiplay services, and high-volume Internet data center internetworking.
“Edge devices are routers, routing switches, integrated access devices (IADs), multiplexers, and a variety of metropolitan area network (MAN) and wide area network (WAN) access devices that provide entry points into enterprise or service provider core networks. Edge devices also provide connections into carrier and service provider networks.”
In general, edge devices are nothing more than routers that provide authenticated access from wherever you are (most commonly PPPoA and PPPoE) to faster, more efficient backbone, and core networks; in this case to the public cloud provider's network.
The point is that whenever the customer is choosing to connect the provider's edge networking equipment, the connection type must be accounted for . The networking services that you typically see here are as would be expected:
- Routing – here we see protocols such as Open Shortest Path First (OSPF) or Multiprotocol Label Switching (MPLS) for reliability and scalability, allowing edge routers to have redundant links to the core network.
- Firewall – to handle security
- Load balancing – with larger providers we would expect to see multiple layers of load balancing.
- Quality of service (QoS) to manage a variety of traffic types.
Once in the public cloud provider's network, or data center really, we find the same sort of physical, logical switching, and other services that we would expect to find in terrestrial data centers. In addition, we have the cloud infrastructure that manages users, workloads, servers, storage, and the hypervisor components that make up the cloud.
At this point, the public cloud provider will need to connect to a backbone, which is a computer network infrastructure that interconnects various pieces of network, providing a path for the exchange of information and access that would tie all of the provider's locations together.
The public cloud provider would also provide access, likely remotely, to their operations folks for the management, administration and monitoring of their customers, as well as the administration, monitoring, tooling and support of their systems.
Are you using the Public Cloud? What are your experiences? Let me know.