Reading most comments here I feel like nobody wants to give a tiny bit of credit to either Twitter or Google, which I find disenchanting.
It is true that any cloud provider would get good publicity by being able to say that Twitter runs on their cloud, but that is precisely the same reason why we shouldn't just shrug this off as a publicity/business decision, because just like Google might have tried to persuade Twitter other cloud providers will most certainly have tried their best too. If Twitter still went with GCP then I think it is only fair to assume that there must have been some real advantage in going with GCP. I don't think anyone who says this must have been a pure business decision is entirely honest with themselves, because a cloud which cannot handle Twitter's volume is no good even if it was entirely for free.
I have nothing to do with Twitter (actually don't even like Twitter because it is such a toxic social media platform) and I don't like many things which Google does, but as an engineer myself I can only say that after using Azure, AWS and GCP for many years in a commercial setting that GCP is indeed years ahead. The quality, speed and reliability of GCP is second to none from my experience and I honestly couldn't say the same about AWS and especially not at all about Azure.
I'm not worried about GCP's quality or technical merits.
Google's management is what scares me. I'd never build a business around a company with so little follow-through, commitment to product longevity, or focus.
> The quality, speed and reliability of GCP is second to none from my experience
Can you be more specific on what reliability GCP is years ahead of the others? I'm guessing numbers will be pretty comparable on all platforms so if we want to declare outright winners it would be useful to cite a number to indicate where you experienced this.
the quality, speed and reliability of GCP is second to none?
I hope this is a joke. AS someone who started on GCP, and got tired of the TERRIBLE quality and reliability of their docs among other issues saying that they are ahead is an absolute joke.
I've found some corner cases with AWS, but got quick resolution even on things they have in "beta" - and consistency from documentation through to system is high.
Secondarily, AWS seems to support even old tech FOREVER. I have a simpledb based app. That tech is 9 years old now. Still ticking. When you are building stuff up over time and can't afford to rebuild on the new hotness every other year, this is nice.
Anyone know the revenue AWS and GCP generate? Seriously hard to believe GCP is so many years ahead in this space from my own experience.
It is true that any cloud provider would get good publicity by being able to say that Twitter runs on their cloud, but that is precisely the same reason why we shouldn't just shrug this off as a publicity/business decision, because just like Google might have tried to persuade Twitter other cloud providers will most certainly have tried their best too. If Twitter still went with GCP then I think it is only fair to assume that there must have been some real advantage in going with GCP. I don't think anyone who says this must have been a pure business decision is entirely honest with themselves, because a cloud which cannot handle Twitter's volume is no good even if it was entirely for free.
I have nothing to do with Twitter (actually don't even like Twitter because it is such a toxic social media platform) and I don't like many things which Google does, but as an engineer myself I can only say that after using Azure, AWS and GCP for many years in a commercial setting that GCP is indeed years ahead. The quality, speed and reliability of GCP is second to none from my experience and I honestly couldn't say the same about AWS and especially not at all about Azure.