More things Kubernetes: What else you missed at KubeCon last week
This is the second part of a two-part series. Click here for part 1.
It’s a good thing, therefore, that I interviewed the hottest vendors who would have exhibited at KubeCon last week. Here are the final half-dozen. (* Disclosure below.)
Aspen Mesh: Istio service mesh for the enterprise
Service meshes are an essential part of any Kubernetes deployment, as they abstract network traffic management among microservices while providing critical observability and security functions. Perhaps the most popular open-source service mesh at the present time is Istio.
Its popularity aside, Istio is still immature and, as with most open-source infrastructure technologies, it lacks the support, management and compliance features that enterprises require for production deployment.
Aspen Mesh fills these gaps with its enterprise deployment of Istio. The company offers a hardened distribution of Istio with a policy framework and a configuration and management dashboard.
Perhaps most important, Aspen Mesh ensures the deployed instances of Istio conform to the business intent of policies in the framework, an example of what I call intent-based computing.
Elotl: ‘nodeless’ Kubernetes that abstracts hybrid and multicluster deployments
In Kubernetes-speak, a node is a virtual machine instance (or in some situations, a physical server) that runs one or more Kubernetes pods in a cluster, or perhaps the entire cluster. Ideally, developers deploying applications to Kubernetes don’t have to worry about the underlying nodes. After all, Kubernetes handles the autoscaling of containers, pods and clusters automatically.
In reality, deploying Kubernetes in an IaaS cloud requires engineers to decide on the particular nodes that will run the Kubernetes instance. In AWS, for example, EC2 provides a plethora of node sizes and capabilities. Nobody wants to choose EC2 instances that are too small, so overprovisioning is a common – and expensive – mistake.
Elotl offers a Kubernetes control plane that offers two essential capabilities. First, it presents a virtual worker node that abstracts all of the underlying nodes, giving developers the appearance of a single node that encompasses the entire Kubernetes environment, even when it includes different Kubernetes flavors in multicloud or hybrid scenarios.
Second, Elotl handles the provisioning of the actual nodes under the covers, ensuring that provisioned nodes are the optimal size and configuration to meet the SLA requirements for each application.
Elotl thus eliminates the problems of overprovisioning and fragmented capacity, handles both horizontal and vertical pod autoscaling, and offers hybrid and multicloud Kubernetes abstractions to boot.
Kublr: enterprise-ready container orchestration
The problem with all the managed Kubernetes services from the likes of Amazon, Google and others is that each of them necessarily depends upon a single service provider. As enterprises seek to scale out their Kubernetes deployments, however, they typically want to leverage multi-cloud and hybrid scenarios that support multiple Kubernetes providers.
Running multiple clusters in multiple different flavors of Kubernetes, however, is a management nightmare. To address this problem, Kublr offers its own Kubernetes platform that supports multiple clusters via a simple management interface, even when those clusters run in different clouds or on-premises or both.
The Kublr platform handles the complexity of scaled-out enterprise Kubernetes by managing ingress, storage, networking, management, and application deployment, all via consistent container runtimes. Via its dashboard, operators can deploy Kubernetes clusters and centrally manage them via automatically deployed observability tools.
Spectro Cloud: Kubernetes-as-a-service for public and private clouds and bare metal
Spectro Cloud addresses the problem that enterprises would rather not rely on a single managed Kubernetes offering by providing a fully managed “as-a-service” Kubernetes deployment.
An important differentiator of the Spectro Cloud offering is its flexibility: Organizations can deploy Kubernetes stacks with different configurations for different teams, which can be especially useful for parallel DevOps efforts.
Spectro Cloud is also useful for deploying multi-cluster Kubernetes at scale, since the Spectro Cloud service’s cluster profiles maintain consistent configurations across clusters, allowing “preflight checks” of configurations prior to deployment.
TriggerMesh: cloud-native event-driven integration platform
TriggerMesh provides an event-driven integration platform that can trigger Amazon Web Services’ Lambda and other serverless functions.
It supports a variety of asynchronous integration patterns, including message queueing and publish/subscribe. TriggerMesh thus can integrate legacy on-premises middleware with serverless functions, as well as event-driven calls to asynchronous infrastructure products like Apache Kafka.
TriggerMesh works with other types of triggers as well, for example Oracle database triggers. Use cases can be simple interactions, such as triggering an AI-driven facial recognition algorithm in the cloud, to more complex scenarios, including event-driven data pipelines that connect multiple technology endpoints.
Volterra: platform for distributed cloud services
Volterra offers distributed application management and secure connectivity on top of cloud-native infrastructure in order to help its customers manage distributed applications and data.
Volterra helps organizations manage multicloud scenarios, but its sweet spot is the edge. Edge computing calls for various deployments of cloud technology in a variety of physical data center form factors, from cellphone towers to retail server closets.
The company resolves differences in application programming interfaces, services and observability capabilities across the various cloud platforms, private clouds and edge data centers.
Volterra supports zero-touch edge provisioning similar to many SD-WAN products, and also provides automatic clustering and autoscaling of nodes. In addition, Volterra distributes identity management capabilities to every endpoint.
The six companies in this article are each finding a way to rise to a singular challenge: scaling infrastructure across hybrid (on-premises and cloud) and multicloud environments.
Clearly, enterprises don’t want to put all their eggs into any one cloud – or Kubernetes – basket. But the demand for such complex environments is deeper than that. Sometimes regulatory compliance drives a hybrid information technology strategy, while in other cases, hybrid IT is part of a complex modernization initiative.
Regardless of the motivations, both hybrid IT and multicloud are here to stay, and are thus fundamental to cloud-native computing in general. Kubernetes is not a one-environment tool, and the providers in this article are ensuring that enterprises are able to achieve their scale and agility goals nevertheless.
(* Disclosure: Volterra is an Intellyx customer. None of the other organizations mentioned in this article is an Intellyx customer.)
Jason Bloomberg is founder and president of the analyst firm Intellyx, which advises companies on their digital transformation initiatives and helps suppliers communicate their agility stories. None of the organizations mentioned in this article is an Intellyx customer.
Image: Linux Foundation
A message from John Furrier, co-founder of SiliconANGLE:
Your vote of support is important to us and it helps us keep the content FREE.
One click below supports our mission to provide free, deep, and relevant content.
Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.