Cloud Native Computing Foundation KubeCon + CloudNativeCon Europe 2020 - Virtual, 4 Sep 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Sponsored Keynote: Cloud Native is Edge Native Too! - Dr. Thomas Di Giacomo

Description

Don’t miss out! Join us at our upcoming events: EnvoyCon Virtual on October 15 and KubeCon + CloudNativeCon North America 2020 Virtual from November 17-20. Learn more at https://kubecon.io. The conferences feature presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects.

Sponsored Keynote: Cloud Native is Edge Native Too! - Dr. Thomas Di Giacomo, President of Engineering and Innovation, SUSE

We all know that the cloud is a great place to run cloud native applications, but that’s really just the beginning of the story. Cloud native is as much about how we deliver applications as it is about where we run them, whether that’s in the datacenter, in the cloud, or at the edge. In fact, edge computing is one area where cloud native technologies can be, and are being, especially well applied. In this talk I will highlight some exciting work being done in CNCF community projects to adapt cloud native technologies for edge use cases, including how they connect edge and cloud to enable edge computing solutions. With these advancements we can employ cloud native technologies to work in ways that are not merely cloud agnostic, but are also edge agnostic.

https://sched.co/ZfHL

A

Hi everyone and thank you for joining this cloud native, is edge native to short discussion. I'm thomas, the giacomo from souza and the one thing I both cannot ignore and cannot talk about today is souza and rancho the chameleon and the cow or the so-called chameleon.

A

Now that it's out of the way, let me share with c-cloud native technologies and many of them coming from the cncf community. Efforts are based on them, how they are applicable to edge use cases edge. Computing has a lot of applications and across many different industries. Its adoption keeps growing similarly to cloud computing edge.

A

Computing requires infrastructure, application and data management and, above all, a smooth and tight integration with on-prem private data centers and public clouds, and while the as a service approach of clouds and the device-based perspective of edge may initially seem very different, they are actually not disconnected to be successful. They are co-dependent on one another.

A

So, let's take a concrete example with smart, almost smart enough to be self-driving cars and bear with me, the picture of the forest shall make sense very soon as an edge and cloud visualization.

A

So, even though there's a lot of compute units and sensors inside a car cows can be seen as the endpoints of such edge architecture.

A

Cars in that forest are the leaves, and they happen to be green because they are smart, so they are clean as well and smart cars to not only collect sunshine there. They are generating tons of data.

A

Some of it is processed locally close to the compute, as, for instance, latency is not a very good friend of driving reaction time. Some of the data is going back to the branches of the trees going back to the on-prem or co-located data centers of the automotive manufacturers, where it is stored, analyzed, processed and filtered.

A

This is a two-way street where the leaves get updates and instructions from the branches too, and actually that's a multiple way streets with exchanges coming from and to the trunks as well and trunks here are public clouds hosting centralized data and user applications connecting with and to the car.

A

Last but not least, and I would have even argued that it's the most important part of the entire forest is the ground, the underground.

A

That's where the seeds are fertilizing and supporting the health of this entire ecosystem, and that's basically, you users and companies, I.t and developer professionals and teams, tools and processes to manage applications, infrastructure, security, observability and so on. This is the layer that makes the entire forest a possible reality, so cloud native technologies are very relevant for edge use cases, either directly or with some adaptations, and in any case they help connect the dots with compute, for instance, as linux and containers, can virtually run anywhere.

A

To illustrate that, let me share just a few examples and there's many more of some of the projects and where they could sit in that forest, as each layer has different purpose and different requirements.

A

Our leaves the edges can be anything from sensors gateways to small data, centers, usually there's a lot more leaves than branches and trunks and they're also much smaller. Hence the need for optimized lightweight solutions, so technologies like cube, edge or k3s, enable edge-friendly, kubernetes environments or long-horn edge-friendly storage, persistent storage, back-end.

A

The specific needs at that level apply to anything networking and connectivity, as well as the operating systems, with special flavors of linux, for edge technologies. Running on specific hardware and chipsets are often needed at that level as well and arm is an example of that going to the branches or on-prem data centers, that's where you have a little more resources, more traditional computes and infrastructure setups.

A

This is also where rook with safe would be a very relevant storage backend, and it is where full-fledged kubernetes distributions and clusters can sit too and then to the trunks. We have the services, public clouds offer there say aks, eks, gke and others all of that has to come together.

A

If anyone becomes unhealthy not implemented well, not talking to the others properly. The entire forest is at risk on the ground. We need to abstract this whole ecosystem as much as possible. Yet we need to manage all the components and all the layers of that forest which are living and evolving things and, of course, edge. Computing networks get very fast, very large, very fast, and they include very different types of clouds and devices.

A

Thankfully, there are projects that give us the mean to cope with all the layers via multi-cloud, multi-cluster and edge management capabilities.

A

So, in closing, I'd like to remind you that first across industries and use cases edge computing does require an integrated infrastructure that spans from edge to core to cloud.

A

Second cloud native technologies are now being adapted and used for edge computing itself and, more importantly, these advances will enable kubernetes and cloud native technologies to become the unifying platform that make edge computing available to the world.

A

I hope you take advantage of your time here at kubcon to learn more about those projects and others, and that you remember you remember that cloud native is really edge. Native too. Thank you.