Well-known big player now moving into hybrid and multi-cloud business
How could it be otherwise, with Amazon EKS Anywhere the company now offers an official Kubernetes distribution, which is designed on the basis of VMware vSphere. The solution offers various options for the deployment and management of Kubernetes clusters in enterprise data centers.
Before we take a closer look at Amazon EKS Anywhere, let's take a look at the cloud variants:
The"public cloud" is freely available and accessible to everyone via the Internet. These cloud services are provided by various external companies around the world. The three largest public and international cloud providers include Amazon Web Services (AWS), Microsoft Azure and Google Cloud Platform.
The"private cloud" makes its cloud resources available exclusively to certain companies or organizations, and unlike the "public cloud", it is not accessible to the general public.
The"Hybrid Cloud" combines the advantages of the "Private Cloud" and the "Public Cloud" and thus offers flexibility and security at the same time.
"Multi-cloud" is the use of multiple cloud computing and storage services in a single heterogeneous architecture.
Open source distribution of Kubernetes
Amazon EKS Anywhere is based on EKS Distro, the open source distribution that helps you create reliable and secure clusters. EKS Anywhere is also open source, but unlike EKS Distro, customers can purchase official support from Amazon if they wish. Because it is managed and supported by AWS, it is a very idiosyncratic distribution. Components such as the Cilium network plugin can be used and not replaced with other plugins. However, CNI-compatible plugins remain freely selectable with the EKS distribution.
Support costs money
Unfortunately, the whole thing is still not quite cheap including support: AWS customers can purchase an EKS Anywhere support subscription for one year at a hefty price of $24,000 per cluster or directly for three years at a price of $18,000 per cluster per year. Another requirement or hurdle is an active Enterprise Support plan, which starts at $15,000 per month. Wow.
At launch, EKS Anywhere supports several plugins offered by AWS partners and the Cloud Native Ecosystem. EKS Anywhere can be installed on a Mac or on Linux to deploy clusters appropriately in development and test environments. It takes advantage of Docker desktop and KIND environment on macOS and Docker Community Engine installations in Ubuntu. However, for active production environments, customers need a VMware vSphere 7.0 or higher to deploy and administer clusters.
Dependencies complicate the use in other environments
Based on the Cluster API interface and the Cluster API provider model, Amazon EKS is dependent on VMware vSphere. This dependency makes it difficult to deploy in other environments, including cloud-based infrastructure services such as Amazon EC2, Google Compute Engine, or Azure VMs. With VMware basically available on the public cloud platforms, it will be very exciting to see if Amazon EKS Anywhere can be deployed in these environments.
Management of external clusters via Connector
EKS Console, the web-based interface originally introduced for managing EKS clusters in the cloud, now also supports registering and managing external clusters running Amazon EKS Anywhere.
Called EKS Connector, the tool can be used to manage external Kubernetes clusters, such as Google Kubernetes Engine or Azure Kubernetes Service. This Connector feature is currently still in the publicly available testing phase.
With Amazon EKS Anywhere, AWS is now entering the hybrid and multi-cloud business as an official big player, competing with products from Google and Microsoft.
You might also be interested in...
Data & Analytics
Data & Analytics serves as a decision-making basis for corporate management and in planning for strategic decisions.
Customer Analytics
The core of customer analytics is to actively support companies in better understanding their customers and their needs.
Predictive Maintenance
Predictive Maintenance and the CoPlanner software to predict company-relevant inventories and manage resources.
Comments
No Comments
Write comment