Amazon: Our cloud growth just sped up. Did you know we are also quite a big retailer? Amazon.com has taken the unusual step of leading its quarterly results webcast with discussion of performance at Amazon Web Services. The e-commerce giant gets most of its revenue, unsurprisingly, from retail sales. But the Amazonian cloud generates much more profit. In this quarter, Amazon's operating income was $14.672 …
I was working at Dell back in 2012, and remember thinking Cloud seems to be a thing, and why, given Dell manufacturers servers, aren’t they building their cloud to compete. The commentary within Dell was around Cloud was too expensive, customers trying it and “repatriating” workloads back on prem and generally putting fingers in ears and saying Cloud wouldn’t take off. HP did try to create Cloud, but it just didn’t really get anywhere and they killed it off.
Seems like they both missed the boat and couldn’t recognise the threat to their business models, or were too unwieldy to respond. It now seems they will suffer from a slow decline and eventual death or consultation, selling of parts of the business that can’t withstand the competition. Kind of reminds me of Blockbusters or BlackBerry. But what is perplexing to me, is the meteoric rise of Dells share price, which doesn’t seem to reflect revenue stagnation and now decline.
One reason for AWS's success is Amazon's retail business needed the vast majority of their servers idle most of the time so they would be ready for spikes in demand. The same would be true for anyone else with a large presence on the internet. The biggest spikes in demand for unrelated web content are unlikely to occur at the same time. This creates an incentive for businesses to find a way to share. Amazon already had servers that were both idle and earning money. Dell and HP would have to build up the infrastructure from almost nothing without a huge retail business to fund it.
I think you are right—I hadn't really thought about that, but I guess that's why they created it in the first place; it makes total sense. Still, it seems like they (Dell/HP et al.) really failed to respond in any meaningful way. Their growth has stagnated. HP revenue was 29.14B in 2019, and the same five years later. Almost the same with Dell. Feels like they lack vision
A lot of the problem is people treat cloud similar to what happened with the Vmware explosion of use. They spin services up because they can do so easily and they don't switch things off, let alone scale down. There is still a lot of cloud treated as just another datacentre not used as intended. Also, you have to slim down your support function and that's hard to do. But ... it is more expensive than a well run and large on-prem estate but there aren't that many of those. Most are a sprawl of legacy and tech debt. Very rarely do people calculate the TCO of an operation. There are people and services supporting the support people.
The cloud vision is good; just create apps and that should get easier and faster with AI assistants doing the heavy lifting. The AI coding is still immature but that will be a game changer.
Only problem could be the total reliance on machines. Today those machines mosty just look for answers from original work from humans, but if the original work ceases? Without those leaps of true innovation that we occasionally have, how will we advance? We may get more efficient at doing what we can already do. General AI with true human-like intelligence however is scary! With the ability to communicate and decide in a handful of seconds, if it did decide to take over, it's a done deal.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
