HCI Scares Us... The Storage Admins
I have always been a massive fan of PDAs and have used generations of Palm Pilots from the Palm iiiC to the Palm V and the Tungsten series. Even tried the Handspring platform that tried to disrupt the market then. Seems like a long time ago but it was about 13 years ago when it was a norm to carry a mobile and PDA. I remembered vividly using the first generation of the Handspring Treo, one of the first PDA and phone devices. Though clunky and buggy, I could see the relevance of it all.
It did take a while before the Treo before iPhones changed the landscape forever. Nobody would have imagined how 13 years on, we literally do everything off that same mobile PDA (smartphone) platform. I think you can probably tell where I'm going with this.
Traditional storage and the 3 tier architecture have been there forever. Many organisations are heavily vested in it, trust it and often it's highly redundant / reliable. So while every one can see and appreciate the value afforded by a hyper-converged solution, much like the Treo, there is still a lot of caution around it's adoption.
You sure or not? So good ah?
Having lived in Asia most of my life, most Asians grew up being extremely conservative. When something good happens or a product does something that we thought was otherwise impossible, it is often dissed off as being too good to be true. The often used Singaporean or Malaysian colloquial expression would be " you sure or not? so good ah? ". While the marketing and pitches make it seem like it is the best thing since sliced bread, some of us would agree that many a time it is not always as claimed. Which makes it doubly hard to always buy in to new tech. Having said that, we do have to agree that it's value is often valid and does changes things for the better. Similarly with hyper-converged, some of us spent a lifetime perfecting the complexities of storage and now we are saying, let the smarts of the software take care of it, a couple of mouse clicks and a noob with no storage experience run it. It all sounds rather risky and impossible.
It is often not magic what makes new technology work, and much of us are able to explain in depth how we are making improvements in those areas. We also know the risk involved for users embarking on it, hence we always look for low risk entry points, gradual adoption strategies and roll back plans, but staying where we are, not embracing new trends and technology is also the surest way of losing relevance.
It ain't broken, don't fix it...
Bad mantra to live by, given that change is the only constant. More so in tech...
The fear of losing relevance is very real. Virtual administrators and storage administrators historically have had very different roles, and with the rise of hyperconverged, one would think traditional storage guys would surely disappear in to the sunset, but nothing could be further from the truth. I have seen from the field many are now embracing and learning the virtualised stack. Coupled with deep appreciation for the intricacies of data management, protection and availability, these transformed storage specialist are now more sought after than ever before. The many instances I have seen virtual admins accidentally rebooting HCI systems (like they were normal hypervisor boxes, forgetting that now it actually has data on it), putting production and backup data on the same storage, cements the fact that despite its simplicity, still requires some appreciation of data management. I would like to believe that occurrence as such should never happen to a skilled storage admin and we haven't even started discussing about backups! I see a future where there will be no division between virtual admins and storage admins, but just a merged skill set of admins (both storage and systems). Staying relevant is just part and parcel of being in tech.
Prioritise what makes sense today
Doing due diligence and comparison between products or offerings are often a challenging exercise in itself. I think we would all agree, finding the perfect product that does everything we need it to (while not impossible), often is pretty elusive. Prioritise what makes sense today and evaluate the support and roadmap given by the vendor moving forward. Would you look at an offering with shrinking presence and relevance in the vendors offering stack? Or would you rather bet your money on an offering that is at the tip of the tongue of every analyst? Most solutions these days offer the flexibility of not being locked down to a specific vendor, so a wrong choice is not often as fatal. End users can always change out as and when it doesn't work out. The lower (cost) of entry, in my opinion, gives us further justification to try out new approaches.
Lucky for all of us, HCI is one such solution.
With that said, hyper-converged infrastructure (HCI) is now turning 10 years old. Where do we stand? It's definitely becoming mainstream and ready for mass adoption in my opinion. All traditional SAN and server vendors, now have an equivalent HCI offering in their portfolio. Is that a real sign that the traditional monolithic storage is being threatened? While HCI may not be suitable in all cases, maybe and just maybe, we are can agree that HCI is indeed here to stay...
