Contactez-nous Suivez-nous sur Twitter En francais English Language

De la Théorie à la pratique

Freely subscribe to our NEWSLETTER

Newsletter FR

Newsletter EN



Juke 2.2 Adds Multicloud Single Cluster Kubernetes Support

July 2019 by Marc Jacob

In December 2018, Juniper Networks acquired HTBASE and its multicloud container product, Juke. Over the last six months, the team has been busy enhancing the innovative multicloud container platform. Today, we’re announcing Juke version 2.2 with integration with Kubernetes and its container storage interface (CSI), as well as new snapshot and scheduler capabilities.

As Juniper Networks continues to simplify our customers’ journey to secure and automated multicloud, Juke’s storage capabilities are a natural complement to the Contrail portfolio.

Contrail provides software-defined abstracted control and automated workflows to network and secure private cloud IaaS stacks. Unlike other SDNs, it also goes well beyond that. Like Juke—within and between clusters atop any private or public cloud infrastructure—Contrail provides, scales and improves Kubernetes networking and security: container network interface (CNI), Ingress, load balancing, microsegmentation and more.

Juke and Contrail can be used separately but they also make the perfect pair: to enable cloud choice, Contrail provides seamless multicloud policy and visibility; Juke now unlocks multicloud storage, freeing applications to access data anywhere and to move between clusters.

What Juke Does

Imagine providing developers an automated, latency-aware, software-defined file system that spans on-premises and cloud resources seamlessly, including AWS & GCP. This is what Juke can do. It allows developers to focus on writing stateful multicloud applications without managing the complex storage details and variables. It also elevates the experience for infrastructure and operations teams, making storage administration easier than ever.

Presented to containers as simple block devices, Juke provides persistent storage while automating the Ops complexity out of DevOps. Developers can focus more efficiently on application development. Architects, administrators and SREs set policies on both compute and storage resources managed by Juke so resources are deployed in line with IT policy attributes: performance, proximity, cost, reliability, GDPR and more.

With Juke, developers will now be able to use compute and storage resources on demand, quickly and easily, while being assured they are cost, performance and policy compliant per company and architectural objectives.

What’s New in Juke Version 2.2

In version 2.2, Juke adds many improvements, including:

Volume snapshots and clone management to boost reliability and enable mobility
Deployment enhancements to make Juke simple to install, allowing developers to quickly become productive
Kubernetes cluster scale out with compute and storage nodes that can now span clouds and sites or stay local to one cluster availability zone

All of this and the ability to easily manage core storage objects for Kubernetes is available from the Juke user interface. Because the Kubernetes integration uniquely enables a single Kubernetes cluster to run spanning on premises and multiple clouds, administrators will optimize resource usage and unlock new use cases, such as:

Automated multicloud resource access to match changing edge compute or proximity requirements: as users move, the workloads can change storage access behavior to follow
Automated multicloud arbitrage for changing day-night usage patterns: for example, use on-premises resources to save cost, but leverage cloud resources when consumption spikes
Multicloud-wide performance and/or cost optimization: move workloads between private and public clouds depending on resource cost.

Imagine an elastic multicloud application that expands and contracts on demand, recognizing when there is a constraint with a given cloud or region and shifting workloads to other clouds to balance.

See previous articles


See next articles