Spectro Cloud, the enabler in modern Kubernetes (K8s) management software, has announced version 3.0 of its Palette platform.
With Palette 3.0, Spectro Cloud is transforming the Kubernetes operating model for IT operations and development teams, enabling them to thrive even as their application infrastructures enter a new phase of complexity and scale, spanning multiple K8s distributions and execution environments from the datacenter and cloud to the edge.
A recent study by Dimensional Research found that the vast majority of organizations expect to increase their use of Kubernetes over the next 12 months, making it their preferred destination for new and existing apps, from a growing number of development teams.
71% say they will have more development teams deploying applications to Kubernetes in the next 12 months.
90% said they will build new applications for Kubernetes in the next 12 months.
79% said they will migrate more existing applications in the next 12 months.
With Kubernetes rapidly evolving from a dev-centric experimental playground to becoming the enterprise-wide platform for production apps, it’s clear that operations and platform engineering teams need to:
Evolve from a few large, shared dev/test clusters to many smaller clusters, in order to better support different dev teams and improve isolation, app availability and security — but without adding excessive cost and complexity.
Consolidate accumulated Kubernetes clusters across diverse distributions and environments in order to simplify management visibility and control — without a painful and disruptive migration.
Support their developers with quick, easy access to the environments and tools they need to innovate on Kubernetes — while also enforcing corporate guardrails and standards to manage risk.
To solve these challenges, the Palette 3.0 release includes three new powerful integrated capabilities:
Nested Clusters: accelerate “time to cluster,” increase utilization, dramatically reduce costs
Ops teams face a difficult challenge when provisioning clusters for different development teams and applications: give each developer or team their own isolated cluster, with all the cost overhead and delays that entails, or operate shared clusters that have potential consequences in terms of availability and security.
With Nested Clusters, there is a better alternative. Nested Clusters is a feature that enables the partitioning of a host Kubernetes cluster and deploying lightweight virtual clusters on top, similar to how virtualization creates logically isolated virtual servers on top of physical machines.
These lightweight virtual clusters are very fast to fire up, which means they’re ideal to give developers the access they need to a ‘sandbox’ environment without delays. They are much more secure than using namespaces. And they look and feel just like a production environment with all the tooling and policies, unlike running a home lab on your laptop. Best of all, they can be oversubscribed on host clusters and torn down on demand, so ops teams can keep cost and complexity under control and maximize utilization.
Palette’s Nested Cluster sandbox feature enables teams to deploy and manage these virtual clusters coupled with Palette’s enterprise-grade orchestration, visibility, day-2 operations, and the fine-grained role-based access control (RBAC) as well as security normally provided for conventional clusters.
Spectro Cloud Nested Clusters can work with any existing CNCF conformant host K8s clusters including distributions such as Spectro Cloud Palette eXtended Kubernetes (PXK), AKS/EKS/GKE, VMware Tanzu, Rancher, and more.
For a deep dive into Nested Clusters and how they can add value in accelerating CI/CD pipelines, Spectro Cloud is running a webinar on November 30, 2022. Full details are available here. To learn more about Nested Clusters, including a step-by-step guide on how to deploy them in a CI/CD pipeline, check out this blog.
Modern fleet management for new and existing environments: unify cluster lifecycle management, access innovation and escape the potential of vendor lock-in
Already 89% of Kubernetes users run multiple distributions in production, with almost all businesses running clusters in multiple clouds and other environments. Yet only 18% say they can manage all their clusters from a single control plane.
Palette has always provided unified control over the clusters it has deployed, regardless of K8s distribution or target environment. Now, it enables operations teams to extend visibility and control over their Kubernetes estate, bringing existing “brownfield” clusters across many different distributions and environments under consistent management. Customers can:
Attach, monitor and manage existing clusters within Palette, importing clusters from VMware Tanzu and PKS, Rancher (RKE), OpenShift, AWS EKS, Azure AKS and Google GKE. This allows operations teams to benefit from Palette’s granular namespace-level visibility over resources, health and cost that may not otherwise be easy to get from other dashboards.
Deploy add-on layers on top of existing clusters (service mesh, logging/monitoring, etc.) and perform day-2 operations like self-healing, backup/restore, GitOps and compliance scans, or configure granular RBAC, across the full stack. Nested Cluster sandboxes can be also deployed, extending the value and providing developers access to tools of their choice on top of any existing Kubernetes.
Provision new clusters with Rancher’s RKE, AWS EKS and Azure AKS distributions or Palette’s PXK and manage their complete day-0 to day-2 lifecycle, including scaling, RBAC, auto-healing, configuration drift control, rolling upgrades, cost management, certification rotation and OS patching.
Automate migration of existing clusters from Rancher’s RKE, AWS EKS, Azure AKS and VMware’s Tanzu environments to Palette’s CNCF-conformant, hardened PXK distribution, with zero downtime and no manual work involved. This gives ops teams a way to escape vendor lock-in from proprietary or legacy distributions and environments that may not have a clear or non-disruptive upgrade path.
Customers can do all this at their own pace, meaning they have a path to consolidate, migrate away from legacy distributions or cloud provider lock-in, and easily scale out the power and flexibility of Nested Cluster sandboxes across existing running clusters.
Palette Dev Engine: accelerate developers’ ability to build and deploy apps to K8s quickly, efficiently and safely
Operations and development teams have a shared goal: maximize app feature velocity. To achieve this, developers need to be able to get access to the infrastructure they need, without having to waste time managing it.
Palette Dev Engine is a new feature set and user experience within Palette that ops teams can provide directly to their development teams. This gives them self-service access to Kubernetes resources, while avoiding burdening ops with trouble tickets and questions or risking infrastructure security or availability.
Palette Dev Engine includes:
App Profiles: a reusable blueprint for app components that can be shared across development teams
App Services: an easy way to incorporate common functions like databases and message buses into an application
Kubernetes Sandbox: an on-demand cluster experience, delivered through Palette’s Nested Cluster capability, that developers can use to deploy the App Profile for testing or small-scale use
Developer Dashboard: a simple view to monitor the health of all their running applications, inspecting for errors, backing up their workloads, adding new parameters, etc.
PDE is designed to free developers from having to be responsible for the “care and feeding” of the whole Kubernetes infrastructure — even to the level of having to maintain a cluster on their laptop. Instead, they get the self-service access they need. PDE is designed so developers can quickly integrate it into their existing workflows and tooling: PDE can integrate via an API with vscode (support for other IDEs coming soon), and authentication occurs via existing credentials using SSO.
Ops teams can configure PDE to the specific needs of their development teams: specifying the maximum size of app sandboxes, pre-loading default app profiles and app services, defining the underlying infrastructure where the nested clusters run, and setting policies and governance around security, access, and other enterprise requirements.
Additionally, they can reduce operating and capital costs through innovative cluster pause-and-resume functionality. With this control framework in place, developers can safely and easily self-serve clusters for their applications, while ops teams can look after the underlying Kubernetes infrastructure.
Palette 3.0 transforms how operations teams manage and deliver application infrastructure
Together, these three announcements of Nested Cluster sandboxes, Fleet Management for Existing Environments and Palette Dev Engine empower enterprise ops teams to streamline their Kubernetes infrastructure and processes even in times of rapid growth: bringing all clusters together, consolidating host clusters through the use of Nested Clusters and giving all their developer teams easy access to sandbox environments and tooling to reduce manual rework when building and deploying all kinds of apps to Kubernetes clusters.
“This is a powerful feature set for our enterprise customers, who are constantly trying to streamline infrastructure operations and improve the experience for their developers, who are ultimately the consumer of Kubernetes,” said Spectro Cloud co-founder and CEO Tenry Fu. “In order to truly enable app development teams to innovate at the speed of business, you have to simplify the developer experience to learn and use Kubernetes. In this major version of Palette 3.0, we have created an environment where dev teams can quickly model and deploy apps using curated services without being Kubernetes experts, while dramatically increasing efficiency and cost benefits for IT.”