> Pretty much every time we picked an unfair - but fast - locking model in the kernel, we ended up regretting it eventually, and had to add fairness.
This is an interesting quote from Linus. I always thought that fairness was for the desktop and Linux had latency issues sometimes because it historically put the "big-iron" first.
Now this quote makes it seem that fairness is a good thing for the server as well and the preceding paragraphs hint that supposed latency issues are just a natural consequence of the trade off between throughput and latency.
This is an interesting quote from Linus. I always thought that fairness was for the desktop and Linux had latency issues sometimes because it historically put the "big-iron" first. Now this quote makes it seem that fairness is a good thing for the server as well and the preceding paragraphs hint that supposed latency issues are just a natural consequence of the trade off between throughput and latency.