Linode Kubernetes Engine (COPY)

Linode Kubernetes Engine (LKE) is Akamai’s managed container orchestration engine built on top of Kubernetes. Through LKE, you can quickly deploy and manage your containerized applications without needing to build and maintain your own Kubernetes cluster. This enables you to utilize Kubernetes without the specialized knowledge, added complexity, and additional overhead typically associated with manual deployments.

Akamai’s managed Kubernetes service is now offered in two tiers: LKE and LKE Enterprise. Each of these tiers accommodates different use cases and deployment sizes. Review the sections below for an overview of each tier and a comparison of their feature sets, resource limits, and capabilities.

  • LKE Enterprise: Designed to be a more robust, scalable, and enterprise-grade Kubernetes management solution for larger deployments. The control plane is highly available and offers dedicated resources.
  • LKE: Designed for smaller deployments and offers a standard array of cluster management features. All LKE clusters are equipped with a fully-managed control plane at no additional cost. Only pay for the add-on features, worker nodes, and any other resources that your application uses (such as NodeBalancers and Block Storage volumes).

LKE Enterprise

LKE Enterprise has recently been released as part of a limited availability beta and offers a robust set of enterprise-grade capabilities. It is more scalable than LKE and can support a larger number of worker nodes and pods. The control plane is highly available by default and offers dedicated resources, reducing possibility of resource contention and increasing the overall performance. Unlike LKE, LKE Enterprise is a paid service. Communication between pods in the cluster is isolated from other traffic.

  • Dedicated HA control plane: The control plane is equipped with dedicated compute resources and is now HA-enabled by default.
  • Increased scalability: LKE Enterprise clusters can support up to 500 nodes and 5000 pods.
  • Enhanced security: Control plane ACLs are enabled by default and Cloud Firewalls are automatically configured on worker nodes.
  • Isolated high-performance pod-to-pod networking: All traffic between pods occurs over the worker node’s VPC interface and is isolated from other traffic within the region.
  • High-performance ingress service networking
  • Enterprise-grade load balancing: When load balancing resources are configured in an LKE Enterprise cluster, they are deployed as Premium NodeBalancers, a more advanced version of our NodeBalancing solution. For more details, see the Premium NodeBalancers documentation.
  • More performant CNI: Instead of Calico (which is offered in LKE clusters), LKE Enterprise uses Cilium as its CNI (container network interface). Cilium is more performant and provides more powerful observability features.

LKE

LKE was designed to make Kubernetes more approachable to developers and small businesses. As such, its feature set is geared towards smaller deployments and focuses on ease of use and simplicity. Container orchestration is handled by the fully-managed control plane. This control plane utilizes shared resources and is not highly available by default (though an HA control plane is available as a paid add-on). LKE itself is a free service, though users still pay for add-ons and resources deployed as part of their LKE clusters, including worker nodes (Compute Instances), load balancers (NodeBalancers), and persistent storage volumes (Block Storage).

  • Free fully-managed control plane: The control plane on a Kubernetes cluster is responsible for managing the cluster's worker nodes, resources, and Pods. Basic control plane infrastructure is provided at no cost. An optional high availability control plane is offered as a paid upgrade.
  • Kubernetes Dashboard: All LKE installations include access to a Kubernetes Dashboard installation.

Features common to both tiers

  • Automatic monitoring, backup, and recovery: A snapshot of your cluster's metadata is backed up continuously and your cluster is automatically restored in the event of a failure. In addition, all of the control plane components are monitored and, if a failure is detected, they will automatically recover.
  • Third-party integration: Harness the strong open source ecosystem of Kubernetes tooling. LKE supports integration with popular K8s-related tools, such as Rancher, Helm, Operators, and more.

Compare LKE Enterprise and LKE

LKE EnterpriseLKE
Maximum # of worker nodes500250
Maximum # of pods50001000
Control plane resourcesDedicatedShared
High availability (HA) control planeYesOptional add-on
VPC isolated networkingYesNo
Node pool autoscalingYesYes
Node pool data encryptionYesYes
Automated Kubernetes version updatesYesYes
Container Network Interface (CNI)CiliumCalico
Load balancing solutionNodeBalancers, Premium NodeBalancersNodeBalancers
Kubernetes Dashboard pre-installedNoYes
Akamai App PlatformComing soonYes
PricingPaid service (see Pricing)No additional cost beyond worker node resources, add-ons, and additional deployed services
AvailabilityLimited availabilityAll core compute regions

Availability

LKE is available across all core compute regions, but is not yet available in distributed compute regions. LKE Enterprise has been released with limited availability and is only available in select regions.

Pricing

For LKE clusters, the basic control plane infrastructure is provided at no additional cost. An upgrade to a high availability control plane is priced starting at $60/month per cluster, depending on region. All other resources consumed by the cluster are billed at the normal rate, including Compute Instances, NodeBalancers, and Block Storage volumes. LKE Enterprise clusters are available to approved customers at $300/month in addition to any resources consumed by the cluster. Review the Pricing page for more information on those costs.

Technical specifications

  • Equipped with a fully-managed control plane. While the control plane is fully-managed, the user is responsible for managing their deployment configuration and applications.
  • Kubernetes version availability: When deploying your cluster, you can select from multiple Kubernetes versions. These are periodically updated with the latest features and security updates. We also regularly make new Kubernetes versions available and remove older versions when they are no longer supported.
  • Compute Instance plans supported: Dedicated CPU, Shared CPU, High Memory, GPU, and Premium CPU (excluding the 512 GB size for both Dedicated CPU and Premium CPU). See the Pricing page for the compute resources included in each plan.
  • 40 Gbps inbound network bandwidth
  • Free inbound network transfer
  • Provisioning and management through Cloud Manager, Linode CLI, or programmatically through the Linode API

Developer resources