Juniper Wants to Simplify Storage Management Across Hybrid Clouds

After simplifying network management with its Contrail products, Juniper Networks now wants to do the same with storage management across hybrid-cloud environments.

Earlier in July, six months after it acquired the startup HTBASE, Juniper shipped an upgraded version of Juke, HTBASE’s flagship product, which makes it easier for enterprises to manage storage across multiple public cloud providers and within their own data centers.

Scott Sneddon, Juniper’s senior director and evangelist for multi-cloud solutions, described Juke as a distributed, persistent storage platform for Kubernetes applications. By simplifying storage administration, it simplifies deployment of cloud-native, multi-cloud applications.

More specifically, it’s a software overlay that allows enterprises to centralize management of data stored in public clouds and in their own data centers by simply setting policies. The software takes care of all the complex underlying details, according to a blog post by Juniper CTO Bikash Koley.

“Juke now unlocks multi-cloud storage, freeing applications to access data anywhere and to move between clusters,” he wrote. “Imagine providing developers an automated, latency-aware, software-defined file system that spans on-premises and cloud resources seamlessly, including AWS (Amazon Web Services) and GCP (Google Cloud Platform). This is what Juke can do. It allows developers to focus on writing stateful multi-cloud applications without managing the complex storage details and variables.”

Related:Juniper CTO Says His Rivals Got Automation All Wrong

Juke works well as a standalone product as well as with the Contrail product family, Sneddon said. The company plans to build tight integration between Juke and its Contrail Enterprise Multicloud software-defined networking (SDN) software for companies that want a more turnkey solution, he said.

IDC analyst Brad Casemore said Juniper’s strategy makes sense as the company attempts to capture a bigger piece of the SDN market, which grew to more than $1.5 billion in 2018 and is expected to reach $3.7 billion by 2022.

Contrail SDN ranks third, with 9.3 percent market share, behind VMware, which dominates with 54.7 percent of the market, and Cisco, which ranks second with 18.1 percent market share in the 2018 fourth quarter, according to IDC. Behind them are Nokia, Hewlett Packard Enterprise, Big Switch, and Pluribus Networks, each of them with a market share in the single digits.

Casemore said the SDN market has room for three to four profitable players. Much of the future growth in the market will be driven by enterprises that want to extend SDN into multi-cloud environments, but they face the complex challenge of managing all the disparate cloud resources, he said. By integrating Juke’s storage management features, Juniper will strengthen Contrail and make it more attractive to those customers.

For example, companies can use Juke to keep two copies of data in both AWS and GCP without having to write to the cloud providers’ APIs. If they have data sovereignty requirements to comply with, they can simply apply policies, and it will automatically ensure data is stored in the right locations.

“Juniper sees an opportunity for growth here,” Casemore said. “They are playing a long game to tell a story that they are a very flexible, programmable, and simple alternative to maybe what some of the larger vendors are doing.”

Juniper’s Juke

The newest Juke version 2.2 supports AWS and GCP, but it will support other cloud providers in the future, Sneddon said.

One use case for Juke is bringing applications and data closer to end users users. For example, to improve application performance for remote workers, a US-based company may migrate a containerized application and its data closer to its sales team in Asia by moving it to a AWS data center in that region.

“It’s closer to users, it reduces latency and improves performance for them,” Sneddon said. 

While it’s often easy to move an application, moving the data it uses is usually difficult. Now, US-based companies can use Juke to make a copy of the data from its data center in the US and store it in an AWS data center in Asia. As the sales team uses the application, the data is stored in the AWS data center and then replicated back to the company’s US-based data center, so the data is always synchronized, Sneddon said.

Juke allows companies to take advantage of the portability and scale-up and scale-down nature of cloud-native applications by allowing for the data to be stored persistently and statefully in whatever location it’s needed, he said.