Amazon has a great service, but when I experiment with things on AWS the cost always seems to end up at $30-$100 to run the experiment on AWS where when I use Google they typically send me a bill for $1.48 or more recently I got a $0.24 bill. It always seems more expensive to even kick the tires on something with AWS versus GCP.<p>Having said that the company I work for now was spending so much on AWS they had to do something to reverse the trend. We changed how we provision and deploy services on AWS and it has brought our costs down approximately $75,000 / mo. I don't know what we were paying before versus now though so I'm not sure how much we saved as a percentage.
I think it is an interesting point but it doesn't really compel me to use GCE over AWS. Now if he had written, "here is our abstraction library and we run on <i>both</i> AWS and GCE so switching to either one is seamless." that would have been a good example of containing the risk of having to switch vendors.<p>The argument is much better if you say, "While I'm using vendor X, I'm putting together the technology to use other vendors easily so that I won't be a victim of sudden switching costs." And I suppose that would also be followed by "Our company isn't based on data from one company I can get no where else."
> With all the important things being equal, I choose the service with the much smaller market share and encourage others to do so as well.<p>The problem with this is it increases your risk that the company will simply go out of business, which could be even worse than raising prices 10x as in the example.<p>I think it can also raise risk that the company will significantly change it's pricing model too, they don't want to be a 3% market share.
A financial company I used to work for had a legacy software product used by banks, that the banks themselves had estimated would cost something on the order of $50 million to replace.<p>The product licensing fees reflected that fact and it was basically a cash cow for the company.
Sounds like our experience when Tutum, now "Docker Cloud" decided to go out of beta with a pricing model that punished users with many containers, and little data/usage. Like startups..<p>We jumped to Rancher.
I don't understand the correlation between the opening of the story and AWS vs. GCE at all.<p>Establishing a long-term dependency on "Scaling R Us" was the problem, not Amazon. If you diversify AWS with GCE so that you have established choices, that's cool -- but "Scaling R Us" is still a problem.
I wish more web services provided opensource versions (Service-as-a-Software). Large orgs would still pay for service contracts even if they self host, and just having the option makes everyone more likely to invest on top of your service.
I wonder if and how you plan this economically, like adding a future cost for each hour of work sunk into the platform. Where you eventually reach a margin where it's no longer economical to sunk more hours into the irreplaceable.