Data Storage's Future: Disruption Ahead - Page 2


Want the latest storage insights?

Download the authoritative guide: Enterprise Data Storage 2018: Optimizing Your Storage Infrastructure

Share it on Twitter  
Share it on Facebook  
Share it on Google+
Share it on Linked in  

Software defined storage brings other headaches to established vendors beside the fact that it opens the door to storage start-ups and companies like Red Hat and VMware who are already well recognized outside the storage arena.

One of these is that it can lead to "performance convergence."  If all storage systems are based on commodity hardware – including Intel Xeon multicore processors rather than painstakingly developed proprietary ASICs – and all are introducing SSDs into the mix, then it gets harder to produce a product that offers storage performance that is significantly better than competitors. That means it is harder for vendors to justify selling systems at a premium price unless they can offer innovative and useful storage features that other vendors don't.

Future Impacted by ‘Compressed Differentiation’

But here's the problem: vendors are increasingly suffering from what Chandrasekaran terms "compressed differentiation." What he means by this is that previously vendors could introduce new features into their storage systems and enjoy a lead over rivals of a couple of years before they caught up, the increasing importance of software means that vendors can only hope for a three or four months lead before competitors catch up and introduce similar – or even superior – features.

That being the case, there is little incentive for storage users to switch vendors to get "killer" storage features. They may as well just sit it out for a few months for their current vendor to catch up.

So what does all this mean for established storage vendors? "The traditional SAN has a finite future," says Mark Peters, an analyst at Enterprise Strategy Group. "How long will it be before most data is stored centrally - maybe in the cloud? We will see that in under a decade."

At a Gartner IT Operations and Datacenter Summit held in Sydney in May, Chandrasekaran outlined a number of possible outcomes for the longer term future of the storage industry including:

· Market share leaders losing the most

· Large portfolio vendors giving up

· Smaller companies getting crushed

· Smaller companies becoming more agile, and being perceived as delivering more value

While it's too early to tell which of these scenarios will play out, he is a little more optimistic about the future of established vendors in the short term. "I think that SANs will still be around in five years' time for tier 1 workloads, and many customers will continue to invest in them," he says. "But the vendors will struggle to get more than single digit growth and in some segments it will be close to zero."

When it comes to non-mission-critical workloads, the outlook is bleaker for established vendors. Storage professionals are conservative so any shifts won't happen quickly, but as they gain confidence in cloud-based solutions and other open source systems, the shift away from big ticket proprietary systems and towards open source and cloud solutions will accelerate.

The good news for vendors is that there are barriers to exit from their ecosystems, and there are reasons that customers may stick with them. "Big vendors have relationships with key enterprise ISVs and a large proportion of storage purchases are done by workloads," Chandrasekaran points out. "So vendors will try to maintain relevance with workload-level differentiation."

There's also a good reason for systems vendors who are involved in storage to stick with it. "Systems vendors can't think of storage in isolation," he says. "They need to compete (in the storage space) because if they don't have a storage offering they have no converged story or strategy."

But as margins are impacted – which they surely will be – some big vendors may have to decide whether there is a long term future in their business at all.

Photo courtesy of Shutterstock.

Submit a Comment


People are discussing this article with 0 comment(s)