Container fanatics are proffering both solid grains of truth and magical-thinking bubbles to the masses. Operating on-premises and in public cloud? No problem — put them in containers, the virtualized method for running distributed software, and they’ll go anywhere. Legacy applications need a cloud makeover? Containers make them good as new.

Sifting fact from fiction in such claims, it becomes clear that not all containerized apps are the same. They won’t deliver all that’s expected of them without the right supporting actors. The infrastructure around containerized apps plays an important role. A containerized legacy app with legacy storage is at a lower level of the game than a containerized microservice-based app with containerized storage.

That’s right — containerized storage is now a reality. It helps people get more of what they generally seek from containerization, according to Ori Bendori (pictured, right), chief executive officer of Reduxio Systems Inc., an information technology and computer storage company. That includes cross-environment portability, microservice-based functioning, and affinity with the Kubernetes container-orchestration platform.

Bendori and Jacob Cherian (pictured, left), chief marketing officer and vice president of product strategy at Reduxio, spoke with Stu Miniman (@stu), host of theCUBE, SiliconANGLE Media’s mobile livestreaming studio, and guest host and cloud economist Corey Quinn (@QuinnyPig) during the KubeCon + CloudNativeCon event in Barcelona, Spain. They discussed the difference containerized storage makes in realizing the benefits of containers (see the full interview with transcript here). (* Disclosure below.)

This week, theCUBE spotlights Reduxio in our Startup of the Week feature.

Everyone all together: ‘Containers aren’t VMs’

The first wave of solutions around containers basically positioned them as mini virtual machines, according to Bendori. In this school of thought, users just packed stuff into containers instead of VMs. They then went on their way, just like VMs, to work with infrastructure built with VMs in mind. Users might have achieved a decent bump in performance over VMs.

This trend toward replacing VMs with containers to cut spend and modernize apps appears to be going strong. Forty-four percent of IT leaders plan to replace at least some virtual machines with containers, according to a survey by Diamanti Inc.

But if users simply swap one for the other without some additional infrastructure changes, they won’t extract the real value from containers, Bendori pointed out. “We think you need to go all the way,” he stated.

Going all the way entails new microservices-based infrastructure that is built for containerized apps. Reduxio’s technology is built out of microservices, and this confers a huge advantage in a number of key areas. Users generally won’t get the benefits they seek from containers without supporting microservice-based technologies, Bendori added.

Old storage is a drag in Kubernetes

The degrees of container and Kubernetes “purity” are apparent across the range of companies adopting the technologies.

“We’ve seen some really crazy uses of Kubernetes, where they’re on Kubernetes, but they’re not really … Kube native,” said Daniel Berg, distinguished engineer, IBM Cloud Kubernetes Service and Istio, at IBM Corp, who spoke to theCUBE in December. When companies lift and shift legacy applications to Kubernetes without adhering to Kubernetes principles and practices — like the proper probes, scheduling hints and quotas — it’s difficult to see why they’re on Kubernetes at all, according to Berg.

Storage is one element that might prevent some containerized apps from being “Kube Native.” Containers and Kubernetes have traditionally flailed along in a somewhat troubled relationship. Containers are ephemeral — they come into being and die rather rapidly. Persistent storage is quite the opposite in design and behavior; it’s not compatible with something cyclically created and destroyed. Kubernetes does not support storing state — and most production applications are stateful and demand external storage.

Reduxio’s microservices-based storage architecture for stateful container applications is Kubernetes based and container-native. Its containerized storage is fully at home in Kubernetes, according to Cherian. It features the Kubernetes Container Storage Interface to deliver persistent volumes to pods that run within Kubernetes. But it goes further than this. Any provider with a CSI plugin might claim to be container-native, since they can attach to Kubernetes. Reduxio is in fact fully container-native, and this is a substantial differentiation, Cherian added.

“The important attribute for container native is that it runs within Kubernetes, it’s implemented as containers, and it is orchestrated and scales with Kubernetes. It should not be something that’s separate,” Cherian said. Once users move containers to Kubernetes, they have a true cloud environment where all work runs inside the Kubernetes cluster, he explained.

Reduxio gives storage the fluidity and flexibility of a service; that is  largely what cloud native, containers and microservices are all about. “This is realization of IT as code, where infrastructure is shared, physical resources are shared and your networking, your applications and storage are just services that run on top of physical infrastructure,” Cherian said.

What about the data?

Containerized storage addresses one of containers’ dirtiest secrets: Containerized apps don’t automatically scoop up data and bring it along for the ride. Proponents often trumpet containers’ ability to cross environments, drowning out this crucial fact.

Containers and Kubernetes do make app portability possible, and this is a practical piece to the hybrid IT puzzle. But we must address the question around data portability. “If you move your application from on-prem to the public cloud, data is not with you, because storage is not with you,” Bendori said. 

Reduxio’s unique IP offers an application-based solution to this problem. “When you move [the app], we are moving the kind of metadata we need, which takes a minute or two, and you can start working immediately in the new location,” Bendori said. “If you want to run multiple clouds, both on-prem and public, you would like to have the ability to move stuff quickly. It cannot be that you move the application and a week later the data arrives. It just doesn’t work.” 

Reduxio also prioritizes data, moving hot data first and cold data later. This is important in hybrid operations, Bendori added. 

The storage revolution will be containerized

Reduxio classifies its offering as “enterprise grade.” It offers a full set of data management and storage capabilities, Bendori said. “We believe containers are moving to production … real serious enterprise applications, [and] you need these kinds of capabilities,” he said. 

Reduxio is geared more towards modernized apps rather than legacy ones already tied to old storage systems, he said. These may include CI/CD (continuous integration/continuous delivery), and anything to do with data processing. “I’m not fighting for the application they have had since the 90s,” he said. 

At KubeCon, Reduxio announced it’s started customer evaluations and will soon begin proofs of concept. Its containerized storage product with be available for production by the fall.

Gartner Inc. predicts that by 2022, over 75% of organizations will be running containerized apps in production. A full ecosystem is growing up around the containerized app, and storage must be part of it, Bendori said. Solutions like Reduxio’s are therefore “the future of storage,” he said.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of the KubeCon + CloudNativeCon event. (* Disclosure: Reduxio Systems Inc. sponsored this segment of theCUBE. Neither Reduxio nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE