[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
Re: [servlet-dev] [jakartaee-platform-dev] What do we exactly dislike in Servlet?
|
On 06/09/2022 05:30, Greg Wilkins wrote:
On Mon, 5 Sept 2022, 17:40 Mark Thomas, <markt@xxxxxxxxxx
<mailto:markt@xxxxxxxxxx>> wrote:
> I did some very rough benchmarks in my blog(s) on an early
version of
> Loom: https://webtide.com/do-looms-claims-stack-up-part-1/
<https://webtide.com/do-looms-claims-stack-up-part-1/>
> <https://webtide.com/do-looms-claims-stack-up-part-1/
<https://webtide.com/do-looms-claims-stack-up-part-1/>> But more
current
> and more thorough evaluation is needed.
Thanks for the pointers. I think repeating some of those benchmarks
with
the latest version of Loom would be a good place to start.
In terms of a more thorough evaluation, what did you have in mind? What
sort of use case do you think async would handle better than Loom? I'd
like to examine those to see if we can put some objective numbers on
the
differences and get some feedback from the OpenJDK folks.
I guess the cometd chat test is a fairly reasonable test that has many
mostly idle connections that to hold them in a blocking style involves a
non trivial stack.
However as it supports both async and blocking transports, it may be
that it is optimized for neither. So setting up a similar problem, but
using dedicated loom vs async implementations might give more insight.
I'm on sebatical until October, but will start by reproducing the tests
from my blog once I'm back.
Sounds great. Enjoy your break and please feel free to ignore this
thread until you get back if you want.
I started looking at the tests from your blog yesterday with
MaxDeepKThreads vs MaxDeepVThreads. Initial results are of the same
order of magnitude but the GC performance for Loom seems to be a little
better. That might be down to hardware differences though. I intend to
explore this further and I'll report back what I find.
The Loom API appears to have changed since the blog was published. I'll
submit a PR to update the tests.
One minor question. I couldn't find a license on the test code repo. Did
you have a license in mind>
Cheers,
Mark