This question is really just a port of a similar question on Stack Overflow:

The emerging consensus from that question is that Linux has superior caching to Windows but the specific reasons why have not been pinned down. Does anybody here have any greater insight into the technical explanation?

A few people, myself included, have also noticed that Windows can actually perform better in a VM on top of Linux than it does on the raw iron (native hardware). Why would having Linux host Windows as a client OS increase the Windows performance?

link|improve this question
1  
I am not sure if you mean more aggressive as opposed to better. There are times when you want less caching, not more. Are there objective benchmarks used to test compiler speed that can be run on more than one OS for the same program? – soandos Dec 19 '11 at 22:23
1  
Because it is Not Windows...;-> – Moab Dec 19 '11 at 22:25
2  
I think this is all rather subjective unless you come up with some hard data to back your claims up. – haimg Dec 19 '11 at 22:41
Do you have anything more than anecdotal "we have noticed" comparisons of one being faster than the other? Benchmarks maybe? Otherwise I'd have to say that it may just be in your head.. also different architectures have different overheads and Windows has a lot of legacy support built in. – Mokubai Dec 19 '11 at 22:44
1  
I think that unless someone posts benchmarks so that we have numbers to work with, this is just not constructive. – soandos Dec 20 '11 at 1:34
show 2 more comments
feedback

closed as not constructive by Mokubai, haimg, surfasb, soandos, slhck Dec 20 '11 at 9:34

This question is not a good fit to our Q&A format. We expect answers to generally involve facts, references, or specific expertise; this question will likely solicit opinion, debate, arguments, polling, or extended discussion. See the FAQ for guidance on how to improve it.

Browse other questions tagged or ask your own question.