Have you felt a change in the storage world? Perhaps you have noticed some resignation among industry veterans when considering the future of data storage technology? Perhaps you have found that when it comes to refreshing aging storage arrays, there is less differentiation between competing products – or no exciting new storage technologies. After all, everyone already has a bay with a flash, is not it?
Perhaps there are wars of fiery sellers to regain some excitement. Or maybe many of your essential storage needs are now met by using a cloud service, an unnamed product, or even free software.
For many years, it has been fun to see the big storage providers fighting for a good cause. They used to line up in the front row at big shows like VMworld, competing for the biggest stand to show off their latest products. But since last year, it seems that the importance of storage has declined a few ranks. Of course, market trends, such as software-defined storage or hyper-convergence, have changed the rules of the game in storage. But until then, when the game changed, competitive storage providers adapted. It may be harder to do now that storage is increasingly embedded, integrated, converged, and made more accessible by competitors in the cloud.
Many recent trends involve getting rid of storage as a computer silo, raising questions about the future of data storage systems. How to sell storage with a capital S if no one buys standalone storage? Are we coming to the end of storage as a first-class industry? The answer, to be quick, is no.
Data, full of data …
Data on the storage market show that traditional storage players continue to suffer from shrinking margins for their high-end storage systems. At the same time, the global storage volume continues to grow. With the explosion of data volumes due to the globalization of applications and the emergence of “Web-scale” databases, Big Data analytics, online archiving and the Internet of Things, it is necessary to properly store all these new bytes somewhere.
It’s just not possible to store all this data in cold cloud storage services. If the data are worth keeping, extract as much value as possible. And if it’s important data, you have to govern them, protect them, secure them and finally actively manage them. Even if the storage is in a converged, hyperconverged or cloud-based infrastructure, much of this infrastructure still contains important business data.
Storage is more important than ever and it will be more tomorrow
Many advanced services are stackable, which means they do not rely heavily on storage arrays or lower-level storage components. This allows third-party suppliers to multiply, and while it’s good for consumers, it makes it harder for storage providers whose business is based on higher margin assumptions.
Some advanced data services may benefit from closer integration with low-level storage, of course. But these services and the providers that offer them may find the best opportunity to shine when they converge deeply with more than just storage components. The problem for pure-players of pure storage is that hyperconverged products, which integrate computing, network and other domains, are not confined to the storage silo and take visible market shares to the actors of the storage.
Set the storage to automatic mode
When it comes to the future of data storage technology, perhaps what is in question is more the storage administration than the way the storage capacity is delivered. Is the increased automation of information technology the real culprit? It is possible that traditional enterprise storage – and its requirement for advanced administration expertise – is destined to be supplanted by fully automated components embedded in cloud-based architectures. If so, should the storage experts prepare to close the shop and go home?
Do not get me wrong: IT automation is a good thing. It is at the heart of most new IT initiatives, including orchestration, self-service, and DevOps. Increasing automation is obvious; IT should always strive to automate repetitive, expensive and error-prone tasks. In addition, vendor-validated automation enables many storage management tasks to be integrated into larger systems, effectively masking storage complexity.
There is, however, a ray of light for storage professionals. When an organization writes and deploys an automation script, it creates a technical debt. At some point – or even many times – in the future, this automation will probably have to be optimized, improved, debugged, debugged, profiled or evolved.
The best of all worlds for storage
It may be unpopular to say, but I think that storage expertise needs are evolving into a dynamic operational management style that is closely tied to the DevOps model. We could call this future OpsDev role, in which the infrastructure operator, now agile, dynamically programs and reprograms a computer stack increasingly defined by the software, including associated storage and data management services.
With manufacturers’ support and remote support functions becoming more dynamic and real-time with each release and machine learning algorithms permeating virtually every facet of IT administration, you can argue that Autonomous storage expertise will soon become useless. But no matter how smart remote administration becomes, an on-site professional who knows the business and its data requirements, who knows what’s acceptable to users, and understands the other specific requirements, will always keep corporate responsibility.
We also have big challenges ahead with regard to the future of data storage technologies. Storage may soon be less a matter of choosing every three or five years specific bays with new features than a matter of assembling a set of storage services, as economical, scalable and distributed, deployable to the global scale. And we will have to do all this while delivering the expected performance, instant access to Big Data datasets, meeting on-demand capacity requirements while maintaining complete corporate governance and providing protection and security Data.
Maybe the function will not even be called storage administrator in a few years, but rather something like a chief enabler of data access. Who knows? The fact is that this is really only an nth evolution of storage. Whatever name you give it, storage is more important than ever. And tomorrow, it will still be the case.