Diamanti’s Spektra 3.0 Offers Multitenant Kubernetes Management

1 Sep 2020 1:58pm, by

The general availability release of Diamanti’s Spektra 3.0 today reflects a turning point for the open source software firm as it seeks to cast a wider net for organizations that want to rely on a single platform for transitioning to and managing Kubernetes environments, the company says.

While in the past, Diamanti sought to tailor its software to help its x86 and hardware customers with their software stacks, Diamanti today continues to widen the features it offers for simplifying the management of often disparate multitenant and multicloud cloud native environments and for stateful applications.

“This particular release really signifies a transition that the company is making from our heritage in the past when we were mostly focused on the hardware [compatibility] with our software stack,” Diamanti vice president marketing Jenny Fong told The New Stack. “This release and with the developments we’ve made in 2020 reflect how we’re really transitioning to software-focused features.”

Among the new multicloud management features Diamanti Spektra 3.0 offers, the ability to manage attached clusters ranks high on the list, Fong said.

“Because we’re leveraging from the APIs that are available, we can also attach existing clusters to the management plane,” Fong said.

Diamanti has previously disclosed how Spektra 3.0 is able to help organizations provision clusters on Azure, for example, and to manage clusters on a single centralized management plane. Additionally, the multitenancy features are key, both for enterprises and organizations and for managed service and software as a service (SaaS) providers. With these multitenancy features, it is possible to offer customers or end users within organizations isolated environments. With the centralized control plane, the server provider or operations team can manage updates, installations and other tasks separately for each tenant.

Multitenant use-case scenarios might include a SaaS provider that offers services to various customers with different demands that might involve hybrid clouds and managing multiple Kubernetes clusters distributed across different environments. A company with several subsidiaries might use Spektra 3.0 to delineate and manage the cloud native environments associated with the different companies separated as tenants. In both use cases, the different tenant groups will often have different compliance and regulatory requirements.

“Our tenancy model is starting from a very strict isolation perspective, which is really necessary to achieve some compliance goals. With the managed service provider, it is critical to isolate your literal independent customers from one another,” Brian Waldon, Diamanti vice president of product, told The New Stack. “So, we offering these very strict controls to allow the broadest idea of multitenancy to apply. A lot of the other solutions out in the market treat multitenancy within a cluster, which is really not appropriate if security is the angle that you’re going for with your multitenancy model.”

Other federated Kubernetes-management features Spektra 3.0 offers include:

  • Multicluster management across bare metal and public cloud: Diamanti Spektra will first add support for Microsoft Azure followed by Amazon Web Services (AWS) and Google Cloud Platform.
  • Policy-based multitenancy for enterprises and MSPs: Administrators can organize tenants and projects across multiple managed clusters.
  • Integrated application deployment migration, and disaster recovery for stateful applications: Project members have the ability to deploy and migrate applications and set stateful application policies for other managed clusters through a “single pane of glass.”
  • The release is also an extension of Spektra capabilities as a platform allowing organizations with existing Kubernetes clusters provisioned across, for example, different multicloud and on-premises infrastructures. Spektra allows the organizations to then migrate the workloads onto Kubernetes clusters on-prem or on the cloud-managed with a single interface. In this way, Spektra serves as a migration path or that on “ramp onto Kubernetes,” Waldon said.

Organizations also typically have existing legacy infrastructures and often want to maintain their investments in legacy applications and data centers. One less talked about feature that has existed for about three years is how Spektra allows virtual machines (VMs) from existing legacy environments to run within Spektra’s managed Kubernetes cluster environments. While Spektra is not designed to directly port legacy systems and applications, older or containerized non-Diamanti Infrastructure can also be converted to virtual Diamanti nodes.

“It all comes down to packaging — if it’s packaged as a VM or as a container image” then it can port to Spektra, Waldon said.

Diamanti is a sponsor of The New Stack.

Feature image by Oliver Ivanov on Unsplash.