Hi, 
few more comments:
first, you maybe wonder why another build-in servlet container, as Kohsuke did 
the migration to Jetty more than half year ago. Jakub enrolled his thesis on 
the university at the beginning of October (IIRC same week when Kohsuke 
announced switch to Jetty) and after that it would be quite difficult to 
completely change the topics already approved by the university, so he decided 
to continue with this topic (btw: I wasn't thesis supervisor, just an adviser)

> Do you think your 8 minutes of testing got past the compilation threshold
> of 10,000 method invocations

not sure what Jakub used for testing, but on my laptop I get about 1,500 
iterations per minute, so I would say it's not sufficient for JIT. On the other 
hand, do you think JIT would have any significant impact? IMHO server internal 
stuff (async. request processing, any kind of caching, thread pooling etc.) has 
much bigger impact

> https://github.com/jbartece/undertow4jenkins/blob/master/thesis/tex/projekt.
> pdf?raw=true

FIY: this is not a text of the thesis, I already asked Jakub to write a blog 
post and sum up performance results there


> Is
> the measured speedup mostly related to serving static resources, or
> better general stream handling, or something else?

AFAIK it wasn't investigated into such details - main subject of the thesis 
was migration to Undertow. To have answers for these questions (and really 
reliable comparison), much in-depth analysis and perf. testing needs to be 
done IMHO

Vojta

Attachment: signature.asc
Description: This is a digitally signed message part.

Reply via email to