This article is more than 1 year old

The world was promised 'cloud magic'. So much for that fairy tale

You're paying too much for too much, and the providers know it

Comment Much of the IT community has been willing to tolerate – even encourage – magical thinking about cloud, and plenty of us believed.

I did for a moment a couple of weeks ago when I saw what appeared to be a brilliant demo of drag-and-drop hybrid cloud storage management: from a screen that depicted on-prem and cloud volumes, a couple of clicks and a tug on the mouse shifted petabytes into a cloud, associated the data with applications in both environments, and ported security policy for good measure.

I then asked the inventors of this wonder whether the user moving the data was offered any information about the costs that were just incurred.

The answer: "No. We'll add that soon."

This, and for that matter, any vendor can get away with such pathetic responses because for years we've been told that moving to clouds and adopting elasticity and opex will change everything for the better.

Nevermind that this vendor has bet its future on providing cloudy storage software instead of physical appliances, so the only costs it cares about is what you'll pay to license its code. The pervasive narrative of cloud magic means details don't matter.

Yet the truth has always been that cloud – like any other source of compute and storage resources – tends to sprawl, can be mismanaged, or can fail to deliver on its promise.

Last week, AWS and Microsoft admitted that their customers have realized their cloud costs are out of control. Some of Microsoft's customers may even be ready to walk because the software giant admitted that helping them to control costs will enhance long-term loyalty. AWS blamed a slowdown in growth on customers wanting to get out of commitment to cloudy resources.

Each cloud's admission was made in the context of a quarterly earnings announcement. That makes the confessions all the more important because they represent disclosures to investors that customers have changed behavior in ways likely to have material impact on financial performance.

Officers of listed companies don't casually toss out mentions of potentially revenue-lowering behavior by their customers. Indeed, they're required to reveal that sort of thing so that investors are informed about future prospects.

AWS and Microsoft have therefore informed investors that cloud users are holding their providers to account for letting costs blow out.

Which is exactly what cloud promised it would not deliver. Remember when Microsoft painted itself as the white knight that could save database users from the need for vast and expensive overprovisioning?

Clearly users have figured out that clouds have often taken them for a ride.

I was recently told the tale of a massive cloud repatriation effort, made necessary by a combination of flaky performance and a top two cloud's indifference to finding a fix. The server vendor that picked up an order for 400 boxes displayed an attitude the customer finds very pleasingly different.

A systems architect of my acquaintance also tells me that customers he encounters are nearly always stunned by just how much compute power and storage capacity he can fit into half a rack, and they happily sign up for it on old-school terms because the savings compared to cloud are colossal. Some of the rigs he recommends don't even need a formal datacenter. A couple of well-ventilated cabinets and careful attention to resilient energy and comms connections can suffice. Once orgs do the numbers, their cloud ambitions blow away.

A few vendors are having another try at making cloud optimization pervasive. I'm hearing talk of "smart clouds" and "superclouds."

But clouds are still trying to perpetuate the magical myth. I recently encountered AWS ads screened during a prime-time reality show that show a family happily benefiting from a number of businesses that use the Amazonian cloud, and a cutely precocious child doing likewise.

The ads are really about seeking social license for cloud computing by making it mainstream, relatable, essential – but not only for nerds – and just a little bit magical.

Actual cloud customers now know that cloud magic is just a story. And increasingly, not one with a happy ending. ®

More about

TIP US OFF

Send us news


Other stories you might like