Nearly all major technology companies are saying the same thing. Kubernetes is the next big thing in computing. https://www.marketwatch.com/article...computing-heres-why-51574863351?mod=home-page The Greek word for helmsman or pilot, Kubernetes is accelerating the transition away for legacy client-server technology by making cloud-native software development easier, better and faster. Last week, more than 12,000 developers and executives gathered in San Diego at the largest annual Kubernetes conference called KubeCon. That’s up from just 550 attendees four years ago. The conference goers are all looking for ways to take advantage of Kubernetes and its ability to automatically deploy, manage, and scale software workloads in the cloud. To understand the trend, let’s start with the changing dynamics of software in the cloud. Cloud apps increasingly run in aptly-named containers. The containers hold an application, its settings, and other related instructions. The trick is that these containers aren’t tied down to one piece of hardware and can run nearly anywhere—across different servers and clouds. It’s how Google manages to scale Gmail and Google Maps across a billion-plus users. Alphabet’s (ticker: GOOGL) Google long ago developed software called Borg to orchestrate its in-house containers—spinning them up and down as needed. In 2014, the search giant opted to make a version of Borg open source, calling it Kubernetes. Today, the major cloud providers all offer a Kubernetes option to customers. Aparna Sinha, the director of product for Kubernetes at Google, notes that Kubernetes is built by the same team that created Borg. “We are quite confident in its ability and how it enables applications to run more reliably, more efficiently, and more affordably,” Sinha says. “Kubernetes has really taken off.” Gartner says more than 75% of global companies will run containerized applications by 2022, from less than 30% today. Kubernetes has become the de facto standard for these managing containers. “As enterprises modernize their infrastructure and adopt a hybrid multicloud strategy, we see Kubernetes and containers rapidly emerging as the standard,” Jason McGee, chief technology officer of IBM Cloud Platform, told Barron’s in an email. In terms of who will thrive in the shift to Kubernetes, there are some early leaders. Last month, Microsoft (MSFT) Azure Chief Technology Officer Mark Russinovich told Barron’s he thinks Microsoft’s Kubernetes service is best-of-breed. Some industry analysts are pointing to other companies. When asked for the Kubernetes vendors that came up the most during discussions with customers, Gartner analyst Arun Chandrasekaran listed Amazon Web Services (AMZN), Google Cloud, and IBM (IBM) Red Hat OpenShift. For on-premise companies looking to use multiple clouds, IDC analyst Gary Chen added, “Red Hat right now is the leader in Kubernetes software. They have the early lead.” It is still early in this new big trend. One thing is for sure, get ready to hear a lot more from technology companies on their Kubernetes strategies. The race is on.
all these containers are a pain to deal with. you wont need a sys admin anymore but you need someone that can configure them so they work ... which they never do
The operating system for the data center. It avoids cloud vendor lock-in as it builds a vendor-neutral layer on top of the cloud vendors infrastructure. It's allow you to move work loads between cloud's and give you a better seat and power to negotiate prices with the different vendors. You can take full advantage of price differentiating and run your services where prices cheapest at the moment.
A friend of mine is working for Accenture right now helping large enterprises with their cloud deployments and is heavily involved with Kubernetes. He is banking!
I know Goldman Sachs and Bloomberg is heavy users of Kubernetes. I work in IT and it's very attractive for enterprises and businesses. It give you resiliency, failover, self-healing, horisontal scalabillity, etc. all for free in one package. Old legacy code are supported. Even ancient COBOL stuff can be packaged into Docker container images and deployed/run.