The computing edge is massive. This core truism means that we often work in this space with an inherently disaggregated but essentially connected variety of endpoints, all functioning away in the neural nodes of cyberspace across the Internet of Things, cloud backbone and wider expanses of the known universe.
The nature of the edge space in cloud computing environments gives rise to huge relevance for Kubernetes, the cloud container orchestration technology that has enjoyed almost de facto industry-wide acceptance in recent times. The “problem” with Kubernetes we so often hear about is that it’s tough to use, complex to manage and quite difficult to learn. One firm that has championed an easier route to implementation at this level is Civo.
Jump to:
- What does Civo offer?
- Edge is an extension of cloud
- Kubernetes Fried Chicken
- Why Kubernetes makes edge sense
- Civo’s Kubeflow-as-a-Service
What does Civo offer?
Civo’s cloud-native services offering is based upon K3s, a lightweight Kubernetes distribution. Originally built by Rancher Labs (prior to its acquisition by SUSE), the K3s Kubernetes distribution is said to work well with continuous computing deployments, hence it’s also a natural fit for the always-on edge world of synaptic sensors and signals systems.
“Edge computing is all about providing new flexible pathways to cloud adoption, hybrid cloud and multi-cloud,” said Josh Mesout, chief innovation officer at Civo. “To many, it’s already clear the benefits of having the point of processing closer to the data storage, helping to reduce latency and providing better security.”
Mesout added that in addition to the latency and security factors, the inherent proximity greatly reduces the cost of data transfer between hardware while also increasing the performance, leading to an improved return on investment for an edge compute use case.
“We often find that large problems often benefit greatly from colocation or by close proximity for the data being processed, which is a huge factor for banks — for example, when they consider building their offices close to stock exchanges,” Mesout said.
SEE: The Ultimate Kubernetes & Cloud Certification Training Bundle (TechRepublic Academy)
Edge is an extension of cloud
Mesout suggested that today we can think about edge computing as an extension of the core cloud, be that AWS Outposts, Google Distributed Cloud Edge or in his company’s case CivoStack. The suggestion here is that once any given IT stack is able to consider its edge estate as an extension of its cloud, then hardware can be scaled as required — even if scaling means shipping a new server to the edge location.
What we do know is that Civo’s clusters are known to be incredibly fast, which is something that matters for IoT and edge computing.
“Our customers use Kubernetes everywhere: From OpenShift running on-premises to multi-cluster and cross-cloud setups on AWS, Google Cloud and Azure,” said Natan Yellin, CEO and co-founder of Robusta.dev, a Kubernetes observability software vendor that lists the Hubble Space Telescope among its user base. “In all those environments, I’ve never seen a Kubernetes cluster ‘spin-up’ and then start operating as fast as Civo.”
However, Yellin added that calling Kubernetes “just” a container orchestrator misses the point. In his view, this is akin to calling the first iPhone a touchscreen phone dialer. Why? Kubernetes is now the operating system of the cloud, and it’s changed the way everyone uses the service-based model of computing, storage, analytics and more that the cloud model offers.
Kubernetes Fried Chicken
“Kubernetes is also showing up in surprising places,” Yellin added. “For example, the well-known American takeaway chain Chick-fil-A runs a Kubernetes cluster inside every store. This sounds odd at first, but it’s an extremely smart and solid engineering decision when you think about it.”
Why would Kubernetes matter so much and apply so well to a fried chicken sandwich?
“Because modern fast food chains are incredibly reliant on IT services to function,” Yellin explained. “From point-of-sale devices and digital menu displays to deep analytics about what people are ordering and when: By putting a Kubernetes cluster in each store, everything keeps functioning, even if the internet goes down or AWS has a cloud outage. The chicken will be served!”
Why Kubernetes makes edge sense
Yellin contended that it’s only natural that IoT and edge computing are starting to use Kubernetes too, as Kubernetes is actually built for distributed systems where failures are not only tolerated but positively expected. This makes it a great choice for edge-computing computing use cases.
Civo now aims to take this space one notch louder, principally through a couple of the organization’s major product updates. Newly announced at the company’s first annual Civo Navigate 2023 conference held this February in Tampa is Civo Platform, a technology proposition that represents a fully-blown Platform-as-a-Service offering.
Clearly aligned to offer developers a more flexible and scalable framework for running and developing applications in the cloud, Civo Platform is built in order to shape to and integrate with an individual IT department’s own cloud journey; that is, it can switch from PaaS into a fully-fledged managed Kubernetes service when that point of scale arises.
“Civo Platform includes a ‘Software Bill of Materials’ tool, allowing users to produce a verified record of all components within a software product,” said Civo CEO Mark Boost during his keynote address. “The SBOM has become a vital way for businesses to verify the security credentials behind the development of a piece of software.”
There are recent U.S. government actions that resonate with this level of IT. White House Executive Order 14028 stipulated an SBOM for any work carried out with the federal government, and this type of mandate is being similarly laid down in other advanced democracies.
“Too often developers outgrow their PaaS provider,” Boost said. “We want to be there to grow with businesses, ready to support their changing development needs as they scale up. Developers don’t want complexity: They want an affordable, flexible and customizable approach to PaaS that lets them focus on innovating for the business.”
Civo’s Kubeflow-as-a-Service
Not to miss out on the drive to artificial intelligence and machine learning, Civo also used Navigate 2023 to announce its pleasingly named Kubeflow-as-a-Service. According to Mesout, the surrounding infrastructure required to support ML is both vast and complex.
“As a result, many organizations spend the majority of their time and resources dedicated to ML on setting up infrastructure components, such as process management tools, data verification and machine resource management,” he said.
Kubeflow-as-a-Service is intended to ease that stress factor by running these components as an ML-managed service, supporting the tools and frameworks most required by developers using ML. The company suggested that developers in smaller organizations are often “priced out” of ML due to the economies of scale required to build the surrounding operations and infrastructure. This technology aims to combat that restriction.
“Technological landscapes, either cloud or ML, ultimately evolve when they are democratized and are easily accessible to all,” Mesout said. “We hope to provide every developer looking to use ML with the platform they require to capitalize on its benefits, and in turn, drive the technology forward.”
All of this Kubernetes movement happens concurrently with the Cloud Native Computing Foundation releasing the results of its 2022 Annual Cloud Native Survey. The CNCF says it uses its annual analysis to scrutinize key factors affecting the cloud native community and landscape. The Kubernetes factor is front and center.
“Kubernetes is emerging as the ‘operating system’ of the cloud — Kubernetes has become the platform for running almost anything,” a CNCF spokesperson said. “Auxiliary workloads have outnumbered application workloads (63% vs. 37%) as organizations increasingly adopted advanced Kubernetes platform technologies like security controls, service meshes, messaging systems and observability tools. At the same time, organizations used Kubernetes for a broader range of use cases like build pipelines, scheduled utility workloads and more.”
Has Civo got itself into the right place at the right time with its cloud-native Kubernetes offering? Certainly, the organization has now added fully-fledged PaaS flexibility, ML capabilities and arguably further extended its relevance into the edge computing space, but one imagines that the major three cloud hyperscalers aren’t quite quaking in their boots just yet
Civo’s partner roster appears to be growing by the month and now includes Loft, Dynatrace, Percona, Pantheon and DataStax. While AWS, Google Cloud Service and Microsoft Azure aren’t there, you can bet they’re listening in. The Kube is cool, and it’s getting hotter all the time.
Read next: From start to finish: How to deploy an application with Kubernetes (TechRepublic Premium)