Reimagining storage for a cloud-native world

Very rarely, the “next big thing” arrives fully formed and ready to change the world. Often, it’s a more complicated process. Different pieces of the puzzle emerge at different times, for different reasons. It can take impressive foresight to understand how it will all fit together.

I’d put microservices in that latter category. Since the introduction of containers as a tool of choice for running infrastructure—most certainly, since Kubernetes emerged as the dominant framework for container orchestration—it’s been clear we were looking at the future of software. By breaking down applications into their constituent parts, and disaggregating those parts from OS and hardware dependencies, we gain huge improvements in speed, scale, and agility.

At least, in theory. In practice, while Kubernetes frees developers from having to worry about the complexities of compute resources, there’s one critical area it doesn’t address: storage. In a world generating 2.5 quintillion bytes of data every day, that’s a problem. Fortunately, some of the smartest minds in the industry are working to solve it. We’re pleased to be helping them do it via our recent investment in Portworx. 

Solving data storage at scale

If developers had to manually specify the location, configuration, and requirements of every compute resource a container might need, no one would bother with microservices. It would just be too complicated. But, while Kubernetes does a great job stripping away that complexity for compute, that’s still basically the process developers deal with for storage.

This isn’t a huge issue when you’re working with a few hundred containers. But scale up to hundreds of thousands or millions, which is where the world is headed? And make that storage dynamic, with data moving across multiple clouds and data centers, on premises and around the globe? It just doesn’t work.

Developers can use Kubernetes to schedule a million containers running in the wild in two clicks. But start wrestling with on-premises and cloud storage, and it takes weeks or even months. If microservices are going to scale, developers need a way to connect containerized applications to all the different places data might live, transparently, so they don’t have to think about it. If truly mission-critical applications are going to run on Kubernetes, developers also need to be able to solve storage- and data-related business requirements like data security, disaster recovery, multi-cloud operations and SLA management. Enter Portworx.

Portworx provides an intelligent overlay for orchestrating container storage in a dynamic multi-cloud world. Expose the software to all your storage—block storage, on-premises infrastructure, cloud storage across AWS, Azure, Google, and IBM—and Portworx handles the rest.

Investing in the future

The complexities of stateful applications may seem arcane if you’re not currently working with microservices. Today, that is still a sizable proportion of developers. But looking ahead a few years, microservices will be the dominant software paradigm. There’s just too much speed, flexibility, and cost-efficiency to be gained.

Kubernetes has shown that microservices can work at scale and has quickly become the global standard for managing next-gen compute. I believe that simplifying storage for Kubernetes is a critical requirement. Unshackling data from hardware and software dependencies will help unlock the puzzle that makes microservices take off—and transform the way we develop software.

For Pathfinder, this as an important area of investment for several reasons. First, as more companies’ products come down to software running in the hybrid cloud, meeting the needs of developers will become an important area of focus. If we can make it simple to provision and use storage—so developers don’t have to find a storage admin or break the CI/CD cycle whenever data moves someplace new—that’s a huge win.

More broadly, Portworx’s vision of the future of data greatly resonates with us. We believe in a hybrid world, where every enterprise use both on-premises and cloud resources, often across multiple clouds. HPE is a global leader in storage—and Portworx can help our customers more easily take advantage of it, even as their IT landscape grows more complex.

As the world embraces containers and microservices, we expect to be well positioned for the future of how data is consumed.