Multicloud mania: what to know


When enterprises started moving workloads and applications to the public cloud, it made sense to adapt existing networking technologies to the new domain. But while compute and storage have successfully become ‘cloud-like,’ networking hasn’t.

Cloud networking solutions being offered by companies including Aviatrix, Cisco, and Juniper Network are all vying to help organizations solve networking challenges when transforming their infrastructure to public cloud. But as cloud implementations become more complex, it’s becoming clear that cloud connectivity solutions based on virtualized datacenter networking technologies lack the agility and elasticity required to   build and scale in the public cloud.

To connect workloads from on-premises datacenter servers to VMs in the public cloud—creating hybrid cloud environments—networking vendors introduced virtualized routers to provide connections between datacenter and public cloud resources. This was understandable, as enterprises were already familiar with this routing technology running in their datacenters. Now, however, the hybrid cloud is evolving to the multicloud. At this point, it’s not enough to simply default to the idea of re-creating the ’90s datacenter network in the public cloud.

“Multicloud” is a term with various definitions, but let’s settle on defining it as the use of multiple public cloud infrastructures. An enterprise might start with a couple of applications running on Amazon Web Services (AWS), perhaps evolve to a global transit architecture for hybrid connectivity, then start to add workloads to Microsoft Azure or Google Cloud Platform.

Leave a Reply

Your email address will not be published. Required fields are marked *