Many of you reading this will be familiar with the experience of having to bite your lip as you watch your organisation pursue a ‘cloud first’ agenda in an almost religious manner. You know that the obsession with moving everything into the public cloud as quickly as possible is blinkered and misguided, but you go along with it because everyone seems so committed and you don’t want to look like a naysayer.
This kind of experience came up during a recent briefing with Jeff Denworth, CMO and Co-Founder of VAST Data. If you’re not familiar with VAST, it’s a company focused on the area of extremely high-volume storage management. Its solutions were originally designed to address the needs of customers who want ‘cloud-like’ scalability, flexibility and ease of use, but delivered via an on-prem infrastructure. As Denworth says: “Our customers are generally dealing with upwards of 5PB of data, and you need to think differently when working at this level.”
We’ll get into that ‘need to think differently’ thing in future discussions, but if you’re aching to learn more right now, check out the VAST website, where you can geek-out on the company’s ‘disaggregated and shared everything architecture’.
Have you still time to turn around, or have you gone too far?
Back to the current discussion, we asked Denworth to describe his ideal customer, to which he responded: “Apart from having a need to store and manage data at scale, it’s mostly about timing. Our proposition resonates particularly well with customers that have been aggressively pursuing a cloud migration strategy, have gained enough experience to figure out that it’s not the Nirvana they thought it would be, but haven’t yet got around to letting their storage specialists go.”
As an aside we chatted about cloud evangelists who seem to build their careers by going from company to company, encouraging them to shift everything to the cloud while downsizing internal IT, then moving on to their next job just before all of the problems with an obsessive cloud approach become obvious.
How much this happens in exactly this manner is debatable, but we’ve been tracking the way in which challenges accumulate as the number of cloud services proliferate for over a decade. Put this together with the way in which activity is becoming even more distributed, and many IT teams are seeing the cost, risk and other issues escalate even further.
Cloud should be a means to an end, not a destination
Does this mean that public cloud services are inherently bad news? Of course not, it’s just that cloud adoption should be regarded as a potential means to an end, rather than an end in its own right. Our advice is always to focus on service delivery objectives, and when you do this you generally end up with some kind of hybrid/multi-cloud approach – a topic we’ll be publishing some new research on soon (watch this space).
In the meantime, there’s now enough experience out there to provide the insights necessary to challenge the cloud crusaders and inject a little more rationality into the discussion. And players like VAST and others are consistently demonstrating that it’s nowadays possible to build on-premise systems that can operate reliably, securely and cost-effectively at extreme scale.