Here is a comparison that I'd like to see next: how much time was spent on the following... - porting the application, making all these benchmarks and writing

Here is a comparison that I'd like to see next: how much time was spent on the following... - porting the application, making all these benchmarks and writing a blog post vs - simply getting a VM with more RAM and more CPUs. In the conclusion the author goes on to talk about the options "if I get the budget" vs "without throwing more money at the problem". I'd guess that the cost of the day of a Principal Engineer would far surpass the cost of years running a 4CPU server.