It's like browsers all over again. As long as each provider is offering different things or different implementations of the same thing we will either need to choose one or be forced to go through a lowest common denominator translation layer.
MSFT, GOOG, and AMZN are all smart enough to not turn this into a race to the bottom of who can provide the cheapest scalable instances. They'll all try to differentiate on features and so it's going to be a long time until you can treat these as a utility.
However, I would argue that VPSs are already drop-in replacements and I bet (though I have no proof) that VPSs are the majority of the market. The cloud providers may not have a viable way to provide even the perception of vendor lock-in when it comes to VPSs.
Now it could be that prices and profits stay high anyways as the market is an oligopoly. The U.S. has experienced the same thing with mobile operators. This is also far from obvious though because entrance barriers for VPSs don't seem steep.
These guys make their money off the companies that need to be able to automate infrastructure and that stuff is becoming more, not less, fragmented.
hint hint
It was recently reported that Apple was spending hundreds of millions of dollars on S3 (before they switched). I think it's safe to assume that a couple of billion dollars of AWS revenue is cheap commodity storage.... which has low switching costs.
Amazon has not matched our May 2015 price cut for GCE (they instead have tossed a couple 5% cuts). They responded to our 2014 cut a day later. We saw a similar reaction with Nearline last year; a few months later they introduced S3 Infrequent Access, but it has plenty of caveats (rounding up to 128 KB per object) and still a 25% higher price per byte.
In the meantime we introduced custom shapes for GCE giving our customers a measured 20% savings. Finally, even Microsoft has per-minute billing.
If you're looking for the best price/performance you'll have to come to Google Cloud.
Disclosure: I work on Google Cloud, so of course I want your business ;)
Depends what you're looking for I suppose. If all you need is basic compute instances then, yes, it's a price war. Any products beyond that have a tenuous relationship between the different providers.
Even something as simple as using a columnar DB pretty much locks you into one of the providers because you have to build around all the nuances of SQL Server vs. RedShift (Postgres) vs. BigQuery.
Forget talking about things like infrastructure automation once you're trying to manage this on demand across many environments in a reproducible way.
At some price point, if the prices are too high, it will be cheaper to hire extra people, and get a cage in a few datacenters and some physical hardware.
We have very bursty traffic and run our baseload of ~ 30 severs (mostly m4.larges, with a few smaller utility servers as well as larger servers for back end services) most of the time, but when we need to scale, we ramp up to about 2 - 4 times that, but usually only for a few hours at a time. We also do some batch processing that takes a few high-memory servers for about a day each week.
We monitor costs pretty closely and our monthly AWS spend is still less than the coloc fees, and we are much more scalable now and have multi-datacenter (AZ) redundancy, and by the end of the year, we will have a cold spare in a second region (which will cost us very little since it will be mostly powered off)
We no longer need to vastly over-provision servers to handle our peak load and batch processing needs.
0 - as an anecdote, I recall their salesmen coming to sell my employer their caching solution around 2008. I asked the rep I'd Microsoft was using their caching product on any of Microsofts web properties and he replied that they were not.
https://azure.microsoft.com/en-us/blog/lap-around-azure-redi...
Certain, newer pieces of some of these services, such as the Azure Active Directory service at the heart of Office 365 and some of the newer, complementary Xbox Live services are hosted on Azure.
Granted, that hasn't always paid off (we all know how the fire phone turned out), but the wins have definitely made up for the losses.