Cloud-native is difficult. Actually, it’s not, the saying is borrowed from ‘Life is difficult’, the opening lines of M. Scott Peck’s seminal human psyche analysis book The Road Less Traveled. Cloud actually isn’t inherently difficult at all because so many of the aspects of the Software-as-a-Service (SaaS) model of cloud computing are provided in automated, autonomously controlled – often accelerated and optimized – packages of consumable flexible IT power.
But, if cloud and the drive to build our IT systems with increasing cloud-native applications and services is viewed by the external observer, they might be forgiven for thinking that many of the working mechanics of cloud need a lot of training, upskilling, mentoring and tuition.
The Cloud Native Computing Foundation (CNCF) puts a big emphasis on skills and cloud software engineering excellence. Rightly so, surely? We don’t want the new always-on on-demand world of cloud that feeds the applications and services we all access on our smartphones, tablets and laptops to be clunky and inefficient. After all, we all know what happens when Facebook goes down due to a datacenter server reconfiguration episode, right?
Cloud orchestration means steerage
With skills at the forefront of where CNCF members drive many of their efforts, a good proportion of learnings are being directed towards Kubernetes (pronounced koo-ber-net-ees). This open source software toolset was gifted to the community by Google and works to orchestrate the containers (smaller discrete components of software that exist in cloud networks) used across modern IT systems.
Used in this sense of the word, the term orchestration means Kubernetes can work as a platform to automate the deployment, scaling and operations of database management systems for the cloud. Derived from the Greek word for pilot or helmsperson, Kubernetes is one of the fastest-growing elements of the cloud-native world.
MORE FOR YOU
Among the firms working to try and widen the global cloud-native skills base is CNCF member organization Red Hat. Explaining where his company’s efforts are directed in this space, VP & GM for Red Hat developer tools Mithun Dhar says that his team is focused on technologies that provide Kubernetes services to build ‘interactive computational environments’ that help solve common deployment challenges.
Cloud-native’s common core
The keyword in that last sentence is common. We’re building a lot of cloud and we (the business world and its IT function) are building an increasing amount of cloud-native applications and data services. This means we’re seeing some of the same ‘shapes’, workload requirements, data cleaning tasks, code refactoring, preparation and provisioning processes happen in different places. Where companies like Red Hat can help is to provide an acceleration advantage derived from its wider experiences with customers and its own in-house Intellectual Property (IP), platform strength and expertise.
In this regard, Red Hat is providing ways for cloud-native deployments to be executed more quickly and accurately. Red Hat Advanced Cluster Management for Kubernetes 2.4 provides ‘policy templating’ (a policy being the prescripted way data is managed in any given system) and zero-touch provisioning. Of particular use in edge i.e. Internet of Things (IoT) computing environments, Red Hat also here combines what the company calls ‘validated patterns for edge’ to reduce deployment complexity, save time and improve the accuracy of the cloud-native systems being developed.
Red Hat’s community contribution to Kubernetes learning is called Kube by Example and it features a collection of Kubernetes-focused tutorials, news and community interaction channels. Also keen to push the skilling-up effort is CNCF member Spectro Cloud, which has announced the release of V2 of its Palette platform, the latest iteration of its end-to-end enterprise Kubernetes management platform.
“The ongoing growth of Kubernetes adoption has created an industry need for innovation around how Kubernetes is being used,” says Tenry Fu, CEO and co-founder at Spectro Cloud. “Customers are telling us that there is a gap in terms of ease of use and sophistication when it comes to Kubernetes management, as they prepare to move more container-based applications to production. With this new version, we have created a cohesive set of all the different capabilities that an enterprise needs to confidently scale their efforts. This platform comes with more deployment options to include edge and bare metal support and additional features that provide enhanced cost optimization and governance controls.”
Fu argues that there is no need to choose or balance between usability and flexibility. His firm’s technology attempts to extend and incorporates the vendor-neutral CNCF’s Cluster API for best-in-class declarative Kubernetes management. This mission here for all firms (or so it appears) is a collective goal to make Kubernetes more accessible to all organizations.
Paul Nashawaty, senior analyst at ESG says that the industry is now widely focused on how to move to the next level of Kubernetes maturity and towards production environments.
“This challenge ranked very high in our most recent container studies. There isn’t a large enough talent pool of qualified applicants who are experienced in managing Kubernetes to meet the growing need for management,” said Nashawaty.
Back to school with CNCF
Back at CNCF itself, the organization has created its own Kubernetes Cloud Native Associate (KCNA) exam. Described as a pre-professional certification, this exam is designed for candidates interested in advancing to the professional level through a demonstrated understanding of Kubernetes foundational knowledge and skills.
A certified KCNA student will confirm conceptual knowledge of the entire cloud-native ecosystem, particularly focusing on Kubernetes. It covers how to deploy an application using basic kubectl commands, the architecture of Kubernetes (containers, pods, nodes, clusters), understanding the cloud-native landscape and projects (storage, networking, GitOps, service mesh) and understanding the principles of cloud-native security.
“Training and certification are highly sought by the cloud-native community, with over 100,000 registrations to date for Kubernetes exams,” said Katie Gamanji, ecosystem advocate, CNCF. “Existing certification programmes have been aimed at skilled professionals who have hands-on expertise in production [i.e. working real-world deployed software] environments. Community feedback has revealed the need for a more beginner-friendly and inclusive evaluation, which is relevant to newcomers across different jobs, from engineers to product managers to marketing.”
CNCF executive director Priyanka Sharma has echoed Gamanji’s sentiment and said that training and skills is core to making the cloud-native world really work in live production. In her view, KCNA fills this gap as an entry-level certification that invites new adopters to demonstrate their skills and knowledge of the cloud-native ecosystem
“What really matters in terms of skills development in Kubernetes today is being able to get all software engineers – and other business stakeholders from marketers to data analysts – familiar with (and comfortable with) the complete landscape of the way this technology is used in cloud-native environments. Moving from the basics to actually being to touch and work with Kubernetes code is really empowering. Facts show that cloud-native has become the scaffolding of innovation in the cloud-native world throughout the pandemic that we have all been (and are still, largely) living through. When you want to build technology products and services quickly and also reliably (so that they don’t fall over) – which all organizations are now compelled to do more than ever – you need to adopt the cloud-native paradigm of software development and delivery,” said Sharma.
From what Sharma is saying, there appears to be a lot of job opportunity out there because there is so much demand for cloud-native skillsets among technology vendors and across all types of commercial organizations.
Casting a wide net to gain engineers
Thomas Keenan is senior product marketing manager at Kasten by Veeam, a Kubernetes native backup and disaster recovery company in the ascendancy. As a CNCF supporter and member, Keenan agrees with the wider sentiments expressed in this discussion and analysis of the industry. He reminds us that some reports note that 90% of organizations surveyed are using Kubernetes in some fashion, but, throughout, a lack of Kubernetes training continues to be a major challenge for the community.
“Companies are casting what I would call a ‘wide net’ in order to try and recruit software engineers into Kubernetes administration and management roles in the Americas and elsewhere. Some firms are taking on rookies in cloud virtualization as the next best option, but virtualization skills do not track and translate directly to Kubernetes when it comes to the complexity of real world implementations with their scale and complexity,” said Keenan.
Making note of his firm’s Kasten Kubernetes Learning Series, Keenan points to the usefulness of self-paced labs as a key training vehicle for people entering this sector of the total IT ecosystem. With recent CNCF exam certifications focused on a range of skillsets, from entry-level to advanced, he also advises a long-term view of the skilling-up challenge at hand.
“If basic core Kubernetes skills cover areas like back up, security and looking after application consistency, then we also need to think about roles where real heavy lifting is going on, where big scalability is called for and where hugely complex projects are being shouldered. At the edge in the IoT, we see retail and real estate making increasing use of Kubernetes due to the wide number of data assets and componentized nature of the systems being built to serve this space, especially in the wake of the pandemic. We all need to think about Kubernetes skills now, so this discussion has some legs,” concluded Keenan.
The upshot and ramifications of what is going on here may not quite be clear to us as we stand today. Kubernetes has deep complexity in some areas and isn’t actually hindering the development of cloud in any way at all, but its need for skilled practitioners has come to the fore and should (arguably) be seen as a clarion call for would-be cloud software engineers who want to position themselves well in the job market of tomorrow.
Cloud isn’t difficult, but sequencing the genome of its deeper DNA can throw up what can seem like a triple helix challenge. Get skilled, get connected and get vaccinated.