By Catherine Sbeglia Nin April 17, 2024
Collected at: https://www.rcrwireless.com/20240417/fundamentals/connecting-network-edges-with-a-balanced-multi-cloud-model
For F5, it’s not just about delivering these applications at the edge; it’s also crucial to secure them once they’re there
According to F5’s Senior Solutions Engineering Manager Chas Lesley, prioritizing application location and delivery is “critically important” for multi-cloud interoperability. The smartphone made it possible to have applications at our fingertips, which inspired efforts to get the application as close to the consumer as possible. Applications “can be anywhere,” but the focus now is developing the capability to “really push applications [to] the end of the arm,” he said during a presentation at the Telco Cloud and Edge Form.
F5, he continued, has a “balanced” edge computing model, in which its global network delivers applications with telco tie ins, but it also pushes those workloads out to the customer edge, whether that is remote sites, remote data centers, remote branches or offices. “We can actually push [and] run applications closer to the client, closer to the consumer, wherever they may be,” he said, adding that the balance of centralized control and presence at the edge is “critically important in how [the company] get[s] applications closer to consumers.”
However, for F5, it’s not just about delivering these applications at the edge, but it’s also crucial to secure them once they’re there. “Being able to layer in different security offerings as we deliver those applications, those API endpoints, that enable us to get resources to the edge as we’ve been talking about, is quintessentially important,” he explained. “Be able to offer the same security in an IoT form factor as maybe we were doing in a public cloud connectivity model, if you will, or a telco facility or tier one data center.”
He added that when providing edge security, F5 aims to be “consistent” and deliver analytics and visibility, and driving both through automation. “We want to … make that an automatic journey in terms of being able to deploy quickly, be very agile,” he said.
F5 makes this possible because of what it does in its compute stacks. “This is where we get into SaaS-based controls, what we do in our global network architecture to deliver this security, but also at the customer edge, at the IoT edge, being able to use small form factor hardware appliances that have integrated connectivity services, but being able to run those services in a form factor like a node … in terms of getting resources out to the edge and being able to extend the same level of security services, delivery services that we do in our global network,” he said, explaining further that the company does this at the edge via mesh services, where its networking and security stacks tie in and deliver services, but also offer multi-tenant distributed Kubernetes models to deliver applications in additional to that network connectivity.
He called this model “hugely powerful” and shared that F5 is seeing a lot of traction across in the market from modern manufacturing to medical care facilities.
Finally, Lesley spoke of the importance of ensuring ecosystem visibility across a multi-cloud network. Doing so, he said, enables “pipeline control” for providers as they modernizing their ecosystems. “We’re doing infrastructure as code, we’re doing networking, more networking as code in terms of what we’re looking at in deployment models, being able to tie directly in and automate that fleet of services, if you will, at the edge,” he said.
And while that’s happening, he continued, there should be a “very rich context of telemetry” where it’s possible to not only see how the networks are performing in terms of flows and other performance metrics, but also at the environments from an application performance standpoint. “We talked about the concept of anywhere at the beginning, being able to say, well, I have a fleet of services at maybe my retail edge, or I have a fleet of services that are backhauled at the public cloud, there’s API frameworks or having other connectivity models that support application architectures,” stated Lesley. “Being able to see how the services are responding across those ecosystems is incredibly important.”
Leave a Reply