What point are you trying to get a across? That to optimise a website you need System Engineers? Well that's kind of a given when you are looking at a scale that big.
You will have these exact same problems with MySQL when looking at that scale. You can only scale up to some extent, then you will require techniques to allow you to scale out, like using replication.
If you read the article and look at how there setup is, they obviously are locked into a scenario where they can't even change databases, so the only way they can scale is with hardware. Now you can get more performance out of MySQL but either changing the database engine, or even using a mysql build that has beter performance and is tested. Kinda like twitter. SO the point I am trying to get across is that with ASP.NET and C# the only way to scale is thru hardware, and with other options you can just switch out some software and then you can do hardware scaling. So yea what would you want to do spend cash as a startup on hardware? Or go with this proven software that is free?
No either you didn't read the link or you lack the knowledge to understand what you are reading. Asp.net has data connectivity to a variety of RDBMS including MySql and Oracle. However switching to MySQL wouldn't provide significnatly higher throughput on the same hardware and Oracle for the cost doesn't really make sense for the type of database they need. The only thing which would give significantly better performance is a NO SQL setup like what Google uses but Stack Exchange didn't need that level of performance so the jump in complexity, and design using NO SQL wasn't warranted. Maybe it would be someday if they scaled larger but given their "niche" scope it is unlikely they would ever need that level of performance so the huge code rewrite for NO SQL (not MySql) isn't warranted. The one advantage that MySQL would have is that it easier to scaled out vs scaled up*. Since it is more efficient when deploying SQL Server to scale up vs out that means making good hardware decisions.
Of course we are talking about a scale of 20x to 100x larger than the largest Bitcoin enterprise. The idea that this would be a problem for a startup is kinda laughable (it is a problem most startups wished the had). I would also point out that contrary to common knowledge MySQL is not license free unless the project is open source. As many Bitcoin ventures are closed source they so require a MySQL license.
* Scale up would mean increasing the performance of a single (or small cluster) or database servers. Where scale out would be replicating the database across a much larger cluster to achieve similar performance. Since SQL Server is licensed the licensing costs are lower when scaling up vs scaling out. The drop in server costs at the high end as well as moving storage to the SAN has made scale up less of a critical issue than in the past. RAM has gotten a lot cheaper. Building out a database server with quad xeons (32 cores) and 256GB or RAM as well as high end SAS controller (24x 2.5" backplane) is under $8K. Going to 1TB of RAM, SSL offloading, and off server storage array is still under $10K.